A) Topics for guidance – Identified topics were divided over two categories. 1) Using and interpreting the ENCALS survival model: a) filling in the model and dealing with missing, incorrect or unclear values; b) selecting and interpreting the outcomes; c) communicating the results to the patient; d) uncertainty in estimates of survival; e) timing of prognostic discussion. 2) Individual needs and preferences of people with ALS and their families: a) information needs patient; b) role and needs family; c) patients with severe cognitive impairment and ALS-FTD; d) immigrant patients with a non-western background in the Netherlands.
B) Evidence on patient needs for discussing prognosis in life-limiting disease – A total of 17 studies were included in the review (Fig. 3). Two studies provided evidence on patients with an immigrant background in the Netherlands, 15 studies provided evidence on other patient needs. An additional file contains study characteristics, study findings, and synthesis of findings (Additional file 1). 15 of the 17 studies focused exclusively on patients with advanced, incurable cancer; none of the studies included patients with a neurodegenerative disease.
Based on a synthesis of the evidence, the following themes were identified: tailored information, family support, diverging information needs, and conspiracy of silence. No evidence was found on tailoring discussion to patients with severe cognitive impairments or FTD.
-
Information needs patient. Tailored information. Not all patients want to know their prognosis.[18, 19, 35–37] Information needs differ from patient to patient and prognostic disclosure should be tailored to individual needs.[15–20, 35–37] Asking how much patients want to know, without explaining what information is available and exploring their emotions and concerns, might not be sufficient to elicit their need for information.[15, 19, 37] Some patients want more explicit prognostic information and time frames, whereas others desire a more general indication.[17, 19, 35–37] Some expressed the hope of being on the tail of the (survival) curve,[11, 35] whereas others did not want to hear statistics and time frames fearing that these could potentially cause them distress and threaten their hope.[15, 18] Although patients emphasized that false hope should not be encouraged and uncertainty should be underlined, some patients emphasized the need for physicians to provide hope by indicating positive aspects and good news stories about other patients beating the odds.[11, 17, 21, 35] Furthermore, patients emphasize that physicians should explain that statistics provide group estimates which may not apply to the individual.[7, 18]
-
Role and needs family. Family support. Most patients want to have family members present to provide emotional support during prognostic discussion, but patients said this should be the patient's choice.[16, 17, 36, 38] Diverging information needs. The families’ needs for information can diverge from those of the patient.[16, 19, 35, 37, 38] Even if a patient does not want to know their prognosis, it is possible that their family does want this information which can help them plan for the future and care requirements.[15, 16, 38] In this case, according to patients, the prognosis can be discussed with their family if they want to know and provided the patient has given permission.[19, 37, 38] Although patients and their families might wish to protect each other from bad news, families respected the patients’ right and wish to know.[16, 18]
-
Non-western patients with an immigrant background in the Netherlands. Conspiracy of silence. Families of immigrant patients in the Netherlands, specifically Muslim patients, may prefer to function as an intermediate in prognostic discussion.[39–41] This can result in them maintaining a conspiracy of silence in order to protect the patients’ hope and because of different values and beliefs related to health and dying.[39–41] This can create tensions between the values of Dutch healthcare providers desiring open discussion with the patient.[39] However, this difference in values and the topic of life expectancy can be discussed if done in a culturally sensitive manner.[39–41]
C) Communication guide and consensus procedures – Our recommendations in the communication guide have been divided into three parts (Fig. 1). The first part deals with practical aspects of filling in and interpreting the prediction model, how to deal with missing or incomplete data, uncertainties of the model and estimated survival, how to interpret the results, and which outcomes of the ENCALS survival model to discuss. The second part covers tailoring prognostic discussion to the needs and preferences of individual patients and their families. The third part contains tips on how to provide information on individual life expectancy in stepwise fashion tailored to patient preference, starting with the situation in general (i.e. prognostic groups), and then, if preferred, addressing more specific points (i.e. IQR of the survival probability).
During the consensus procedures the following topics were discussed between the working group and expert panel. First, selection of outcomes of the ENCALS survival model (i.e. prognostic groups, survival curve, survival probability) to discuss and illustrate the estimated life expectancy. Initially, we only considered the prognostic groups and their median ranges to be suitable for this purpose. We assumed that survival curves and survival probability might overwhelm a patient. However, after exploratory discussions of prognosis by members of the working group (MvE, HW, EK), we concluded the IQR of the survival probability to be suitable for illustrating a more individualized estimation. Second, uncertainty in estimates of survival. We recommended that uncertainty of individual disease progress be emphasized by discussing life expectancy as a group range while also pointing out that some patients within this group are better off and others worse off. This can be further illustrated using the interquartile range of the survival probability. Third, timing of prognostic discussion. The working group deliberated whether personalized prognosis could be discussed during diagnosis given the limited time available to fill in the prediction model during consultation, and whether patients would be able to process the information considering the emotional impact of the diagnosis. Due to a lack of evidence, we decided not to make a recommendation on the preferred timing. However, we concluded it would be unethical to continue telling them the average life expectancy without mentioning the possibility of a more personalized prognosis; the option to discuss personalized prognosis, if the patient wants to know, should be offered during diagnosis. Fourth, recommendation on culturally sensitive communication. The working group concluded that offering spiritual assistance while discussing the prognosis is part of core patient-centered communication skills and recommendations on this were not included. We did include recommendations on how to discuss personalized prognosis in a culturally sensitive manner. Fifth, lack of disease insight versus lack of decisional capacity in cognitively impaired patients. The ethicist in our expert panel suggested we should make a more clear distinction between these, since the latter comes with certain patient rights and physician responsibilities. To avoid ambiguity, the working group decided to focus our recommendations specifically on patients lacking decisional capacity to decide whether they want to discuss their life expectancy. An additional recommendation was to use a cognitive screener like the Edinburgh Cognitive and Behavioral ALS Screen (ECAS) to gain insight into affected domains if a lack of decisional capacity is suspected. Finally, the working group discussed whether percentages (50% of patients) or frequencies (2 out 4 patients) should be used to discuss the IQR. We concluded that patients are more likely to understand survival if expressed as a frequency.
In addition to the consensus procedures, a preliminary version of the guide was discussed with rehabilitation physicians working in ALS care during a workshop at the Dutch ALS conference for health professionals (2019). Their comments on filling in the prediction model (including 'conversion' of progressive muscular atrophy (PMA) or primary lateral sclerosis(PLS) to ALS, patient’s country of origin, forced vital capacity upright or supine, and using the model to track disease progression) were incorporated in the text.