We have developed a communication guide to support physicians in discussing personalized prognosis in ALS. Recommendations aim to provide guidance in filling in and interpreting the ENCALS survival model and support physicians in tailoring discussion of personalized prognosis to the individual preferences and needs of people with ALS and their families. Uncertainty in estimation of life expectancy, due to heterogenous individual disease progression as well as inherent limitations of the underlying prediction model, are discussed.[5] Finally, patient choice and the right not to know are emphasized as the basis for prognostic discussion.
Communication of personalized prognosis
Our communication guide focuses on discussing estimated life expectancy based on the ENCALS survival model. Discussion of life expectancy (i.e. quantity) can support the quality of life of patients by aiding patients and their families in decision-making [12, 15, 16] and planning for their care and future [15, 17, 18], as well as providing patients a sense of control.[17, 19] It can also support healthcare professionals in the timing of appropriate and effective care easing the burden of the disease.[25] However, how to provide numerical estimates of survival and associated uncertainties in a manner that supports patient decision-making is a subject of debate.[42, 43] Being too specific can cause distress if survival is underestimated or overestimated,[6] but too wide a range can reduce credibility and accurate understanding.[44] It has, therefore, been argued in oncology and neurology that life expectancy can be discussed effectively using multiple scenarios based on the median and interquartile range to illustrate average survival, and groups worse and better off.[4, 45] This can also help patients prepare for the worst while hoping for the best; a study in cancer patients showed that patients preferred this to simply median survival.[46] Another possible barrier to patient understanding is statistical illiteracy.[47] Visual aids can help facilitate patient understanding,[48] but patients generally prefer words and numbers to graphs and diagrams.[49, 50] Whether estimated survival is communicated visually or in words and numbers, patient understanding can be supported using frequencies instead of single events, absolute rather than relative risk, mortality not survival, and natural frequencies rather than conditional probabilities [47, 51] as we have done in our recommendations.
Non-western patients with an immigrant background
Studies amongst general practitioners and oncologists show that physicians often communicate differently with non-western patients with an immigrant background: consultations are shorter and less focused on involvement and empathy,[52] patients are involved less in decision-making,[53] and more medical jargon is used.[54] However, it is not at all evident that patient needs for prognostic discussion differ between western and non-western patients. Some, but not all, want to know their life expectancy,[18] desire the topic to be discussed first or only with their family,[18, 39, 41] and prefer a more indirect style of communication.[39, 41] Thus, many core skills of patient-centered communication are relevant during intercultural communication.[55] However, one important difference is the role of family. Families of western patients emphasize the importance of respecting the patient’s choice in knowing their prognosis, even though sometimes they would prefer to protect the patient from bad news.[16, 18] Whereas families of non-western patients often prefer to shield the patient from bad news, in order to protect their hope.[18, 39, 41] However, western healthcare values and laws respect patients’ autonomy, including the choice of not wanting to know or letting family make this decision.
Impact of cognitive impairments in discussing prognosis
Cognitive or behavioral changes occur in up to half the patients with ALS,[1] which can impact patient autonomy in, amongst others, decision-making and communication of personalized prognosis. Around 13% of patients with ALS fulfill the criteria for the behavioral variant of FTD,[1, 56] which can cause apathy, reduce insight, and impair decision-making.[57] However, this does not necessarily mean the patient lacks decisional capacity. Therefore, the working group decided to differentiate between cognitive impairment versus a lack of decisional capacity regarding decision-making on discussion of life, and focus our recommendations on the latter. If a lack of decisional capacity is suspected, a cognitive screener can be used to provide insight into affected cognitive domains. A concise screener like the ALS-CBS could be used to screen for behavioral changes; however, a broader screener like the ECAS is recommended because difficulties in decision-making can also be caused by other domains like impaired language or memory.[58] Assessing decisional capacity depends on the physician’s judgement and weighing of multiple relevant factors in addition to cognition (e.g. emotion, motivation, and volition), is specific to the situation, and subject to different legal definitions depending on the country.[57] Discussing estimated survival with the patient’s family, if they want to know, can still be important as they will have to take into account a poorer prognosis due to cognitive impairment.[5, 59]
Generalizability
When using the ENCALS survival model, two limitations have to be taken into consideration. First, although it is becoming more common to consider ALS, PMA, and PLS to be on a spectrum within the same disease,[1] the model has only been validated in patients with ALS.[5] Second, the model has been developed and calibrated with data from 14 ALS centers across 9 countries [5] and can be used to reliably estimate prognosis for these countries using their cohort. Other western countries can use the general ENCALS survival model which can be tailored to regional factors by recalibration of the intercept of the prediction model in future studies. However, the model has not been calibrated for countries in Asia, South-America or Africa, and differences in genetics, healthcare systems, and other factors have thus not been taken into account. An additional consideration is that this guide was developed in the Dutch healthcare setting. However, we believe that our recommendations can be useful to support discussion of personalized prognosis in other western countries. Evidence underlying recommendations, except those on immigrant patients in the Netherlands, comes from international studies and are in line with international guidelines on communication in cancer.[29, 30]
Specificity
While conducting our review, we found no evidence on discussing life expectancy in ALS. Available evidence was mainly based on patients and family caregivers in terminal cancer. It is unclear whether these findings can be generalized to ALS. Whereas in most cancers people are able to retain some hope of being cured, the disease outcome in ALS is homogenous in its invariable lethality and relentless, unavoidable and constant prospect of decline and loss.[60] Possibly as a result, patients with ALS more often engage in advance care planning compared to those with cancer,[61] which can necessitate more information on personalized prognosis. On the other hand, cognitive impairment plays a much more significant role in ALS, even early in disease,[62] which can hinder decision-making and impact decisional capacity,[57] a topic absent from patient-clinician communication guidelines in cancer.[29, 30]
Strengths
This is the first communication guide, as far as we are aware, on tailoring discussion of personalized prognosis in life-limiting disease based on a prediction model. Additional strengths of this project are inventory of topics amongst the target audience, development over multiple rounds of consensus procedures, and feedback by a broad expert panel which included people with ALS and a family member.
Limitations
One limitation of our guide is that the underlying evidence was obtained from studies in patients with terminal cancer and this may not be valid for patients with ALS. A second limitation concerns our search to identify the needs of immigrant patients with a non-western background in the Netherlands. We only found evidence on the needs of Muslim patients with a predominantly Turkish or Moroccan background,[41, 63] the two largest immigrant groups in the Netherlands.[64] However, in formulating our recommendations, the working group and expert panel did take into account all immigrant groups in the Netherlands and our recommendations are in line with Dutch consensus recommendations on palliative care for people with an immigrant background.[65]
Implementation
The ENCALS survival model is accessible to physicians and researchers by registering online.[66] This communication guide is intended to facilitate discussion of personalized prognosis in ALS and will be distributed through the network of the ALS Center Netherlands. In addition, the full Dutch version and an abbreviated English version will be made available online at our website.[67] However, the development of this communication guide is only the first step in the implementation of discussion of personalized prognosis. We are currently conducting a qualitative study to evaluate patient and caregiver experiences with discussing personalized prognosis based on our communication guide. The results of this study will be used to provide recommendations on discussing life expectancy in ALS and the guide will be adapted accordingly.
Conclusion
This communication guide supports physicians in filling in and interpreting the ENCALS survival model while tailoring discussion of personalized prognosis to the individual needs and preferences of people with ALS and their families. Uncertainty of estimated survival and individual disease progression should be emphasized by discussing the estimated life expectancy as a range and underlining that some patients are better off and some worse off. Prognostic discussion should be tailored to individual information needs and preferred level of explicitness. Patients should be given the choice of having family present for emotional support. Families of patients with a non-western background may try to shield the patient from bad news about their prognosis, but, while respecting cultural values, physicians should explain that this is the patient’s choice. When information needs diverge and the patient does not want to know their prognosis, this can be discussed with the family with patient permission. Whether to discuss personalized prognosis or not is always the choice of the patient, including the right not to know. However, if the physician judges that the patient lacks the capacity to make this decision due to severe cognitive impairments or FTD, an exception should be made and life expectancy discussed with their family. An ongoing, qualitative study is currently evaluating the effect of tailored discussion of personalized prognosis on patients with ALS.