Introduction
There have been significant improvements in treatments for childhood cancer, with 5-year survival rates increasing to over 80% in high-income countries (Youlden et al. Reference Youlden, Baade and Moore2022). However, survival for some cancers, including high-grade gliomas and relapsed sarcomas, has not improved over the past 40 years (Youlden et al. Reference Youlden, Baade and Moore2022). These “hard-to-treat” cancers are the most aggressive of all cancers, and, sadly, available treatments often do not work for most children (Youlden et al. Reference Youlden, Baade and Moore2022). For example, diffuse midline gliomas are the most common childhood high-grade gliomas, and almost all children die within an average of 12 months after receiving this diagnosis (Bartels et al. Reference Bartels, Hawkins and Vézina2011). For diagnoses where there are no standard treatments available, patients may be offered an experimental treatment through an early-phase clinical trial or through precision medicine pathways. However, the primary intent of early-phase clinical trials is not curative. The primary aim of Phase 1 trials, for example, is to assess the safety and maximum-tolerated dose of novel drugs and therapies (Berg Reference Berg2007; Hazen et al. Reference Hazen, Zyzanski and Baker2015; Robertson et al. Reference Robertson, Mitchell and Wakefield2019). While many children with very high-risk cancers will have clinical benefits from precision-guided therapies, ultimately, for most patients, these treatments are not curative in the long term. Therefore, alongside any experimental treatments, children may also be offered palliative care to address the physical and psychosocial difficulties that occur alongside cancer-directed treatment.
Healthcare professionals (HCPs) working with children with hard-to-treat cancers are responsible for delivering news of the poor prognosis to families and are required to provide unbiased information and support to families when there may be limited evidence available (Mack and Joffe Reference Mack and Joffe2014). Communication with the family about prognosis and treatment options is typically led by the oncologist and supported by other HCPs, including nurses and social workers (Jones et al. Reference Jones, Currin-Mcculloch and Pelletier2018). Effective communication in oncology requires the exchange of information to enable decision-making, build rapport with families, manage uncertainty, and provide emotional support and validation (Sisk et al. Reference Sisk, Friedrich and Mozersky2020, Reference Sisk, Mack and Ashworth2018). However, research suggests that there are often gaps in communication (Sisk et al. Reference Sisk, Friedrich and Mozersky2020), and parents often report insufficient communication about prognosis (Kaye and Mack Reference Kaye and Mack2013).
A common struggle for HCPs in this field is in balancing honesty about the child’s condition and limited treatment options with parents’ eagerness to “try anything” (Robertson et al. Reference Robertson, Mitchell and Wakefield2019). Literature from adult oncology and life-threatening pediatric illnesses shows that clinicians can report reluctance to disclose prognosis because they worry about the emotional impact of this information on families and do not want to diminish their hope (Gordon and Daugherty Reference Gordon and Daugherty2003; Mack and Joffe Reference Mack and Joffe2014). However, there is little evidence that limiting the delivery of prognostic information preserves hope (Mack et al. Reference Mack, Wolfe and Cook2007) or decreases distress (Marron et al. Reference Marron, Cronin and Kang2018). Effective communication between HCPs, patients, and families may increase families’ satisfaction with healthcare and improve their peace of mind (Sisk et al. Reference Sisk, Mack and Ashworth2018). Furthermore, honest and transparent communication allows families to focus on attainable goals (Sisk and Mack Reference Sisk and Mack2018) in partnership with their primary treating team and palliative care team.
Hard-to-treat childhood cancers require numerous decisions to be made regarding complex treatments, all with uncertain outcomes. With families able to easily access medical information online (much of which is unregulated), HCPs’ responsibility to effectively communicate with families about their child’s prognosis and treatment may be increasingly difficult. As such, the current study explored 3 research questions:
1. What are HCPs’ challenges in communicating with families of children with hard-to-treat cancers?
2. What skills have HCPs developed to communicate with families of children with hard-to-treat cancers?
3. Have HCPs undergone previous training, and what are their attitudes toward formal training, to communicate with families about hard-to-treat cancers?
Methods
Participants
Participants were HCPs (e.g., oncologists, nurses, social workers, and any other relevant professionals) who had direct responsibilities for managing children/adolescents with hard-to-treat cancers in Australia within the previous 24 months.
Procedure
We recruited HCPs via exponential snowball sampling, starting with our clinical investigator (D.Z.). D.Z. emailed study invitations to individuals in his network who were employed at the study site. Participants were then asked to provide other names/email addresses of other potentially eligible participants. We also recruited HCPs via study advertisements on social media. Email invitations and online advertisements contained a link to the information sheet and e-consent form, hosted on REDCap. Once participants provided their e-consent, they were directed to either complete a booking form on REDCap or to email the research team to arrange a time for the interview.
Data collection
We developed a purpose-designed, semi-structured interview schedule that focused on HCPs’ perceived challenges in communicating with families, skills or techniques they had developed to aid communication, and knowledge of, and attitudes toward, training to aid communication. The interviews were used for data collection as part of a larger study that also examined HCPs’ knowledge and access to experimental therapies. Only the data regarding communication is reported in this paper (see Supplementary Material for the interview guide). Interviews were conducted by 4 female psychosocial researchers (S.M., R.D., B.M., and C.M.), who all had training and experience conducting semi-structured interviews in pediatric medicine. Interviews were conducted one-on-one via Zoom or telephone, audio-recorded, and transcribed verbatim. Interviews lasted 38 min on average. Participants were interviewed once, and transcripts were not returned to participants for comment. Given the small population size, some participants had previously participated in our earlier research and thus had known the interviewers before study commencement.
Data analysis
We used qualitative content analysis and an inductive, iterative approach to analyze the data according to our 3 research questions. We followed Elo and Kyngäs’ (Reference Elo and Kyngäs2008) 3 phases of qualitative content analysis, which provides a framework to capture participants’ perspectives in descriptive yet rich categories (Elo and Kyngäs Reference Elo and Kyngäs2008). In the first phase, Preparation, L.K. (first author) became immersed in the data by reading the transcribed interviews multiple times. In the second phase, Organization, L.K. used the guidelines of Miles and Huberman (Reference Miles and Huberman1994) to open-code the transcripts. We used NVivo Release 1.7 (QSR International) to code the data, develop the coding tree, and record the frequency of the codes. S.M. also coded 4 randomly selected transcripts (i.e., 20% of the data) in accordance with accepted recommendations to reduce research bias (Burla et al. Reference Burla, Knierim and Barth2008; Campbell et al. Reference Campbell, Quincy and Osserman2013). L.K. and S.M. then compared coding, discussed and resolved discrepancies, and revised the coding tree. L.K. abstracted the data by organizing the codes into overarching categories and interpreting these categories in a way that accurately captured the original data. L.K., S.M., and E.T. all reread the transcripts, then met to discuss and revise the codes and categories until consensus was reached and to ensure that the themes accurately reflected the data. In the third phase, Reporting, we recorded how frequently we had coded each category and chose quotations from the interviews to represent the categories.
Results
Due to our recruitment strategy, we are unable to determine the overall response rate. However, 32 HCPs opted in to the study, and 25 (78.1%) of those participated. Five participants were then excluded from the study as they did not have direct experience working with patients with hard-to-treat cancers. Participant characteristics are presented in Table 1.
Participant characteristics
a Two participants were not asked this question.
Table 2 shows the frequencies of the categories for the research questions: (1) challenges; (2) skills; and (3) training, alongside representative quotations.
Themes and representative quotations (N = 20)
HCP = healthcare professionals.
Research question 1: challenges
Information provision vs. family distress and need for hope
The most frequently cited challenge was trying to balance providing families with accurate medical information while also managing family distress and maintaining hope. HCPs felt challenged to provide accurate prognostic information to families while not overstating the potential success of various treatment options and creating false hope. Adding to this challenge was the need to cater to the individual needs of each family and each family member, with HCPs reporting that some wanted as much information as possible, while others were not able to process or accept any “bad news.”
Parents want to censor information provided to their child
A particularly challenging aspect of family communication reported by 11 (55%) HCPs occurred when parents decided, on behalf of their child, that their child would be too distressed to hear the reality of their diagnosis and prognosis and requested that the HCPs censor either all or most of the details regarding their cancer. While HCPs reported that this was not a common occurrence, when it did happen, it was a significant challenge to navigate, especially when the patient was an adolescent and capable of understanding what was happening to them. HCPs typically reported that, while their preference was to provide developmentally appropriate and honest information to the young person, they wanted “to be respectful of parental wishes” (Oncologist). One oncologist, however, reported that they would inform adolescent patients of their prognosis and treatments, regardless of the parents’ preferences.
HCPs and families have different expectations of treatment outcomes
Eleven HCPs (55%) reported that a major challenge was having different expectations for treatment outcomes relative to parents. In particular, they shared that HCPs often had “the expectation that it’s unlikely or less likely that there will be benefit derived for that patient” (Oncologist). HCPs shared that some parents were “desperate” to enroll their child on a trial and held immense hope and expectations that it would offer an effective cure. Five HCPs specifically reported that when families accessed information about clinical trials and experimental agents online, it could exacerbate their unrealistic expectations for treatment. One of these HCPs reported that “it’s quite easy for a parent to have distrust in the clinician” (Oncologist) if they found information about new agents via international parent/patient forums that were not recommended by the treating team.
Uncertainty
Seven HCPs (35%) noted that one of the major challenges they faced was uncertainty regarding which treatment option was best, when treatment would begin, whether they could get access to new experimental agents, when those agents would arrive in Australia, and what the outcomes would be. This uncertainty then created difficulties when communicating with families as the HCPs were unable to provide definitive answers to parents’ questions regarding treatment options and the likely efficacy of different treatment options.
Barriers to multidisciplinary teamwork
HCPs reported that multidisciplinary teamwork was preferred when communicating with families about diagnosis, prognosis, and treatment options and addressing families’ resultant information and support needs. Oncologists typically led the conversations with families to deliver “bad news” and noted that ideally nurses and social workers would also be in the room to advocate for families’ needs and provide emotional support. However, 6 HCPs (30%) reported that staff shortages, the busy pace of the hospital, and scheduling conflicts were at times a barrier to having nurses and/or social workers be part of these conversations with families. In addition, all 3 social workers reported that they believed their role was not always understood nor “respected” by the rest of the clinical team, and this led to them being left out of some important conversations with families.
Research question 2: skills
Developmentally appropriate communication
Thirteen HCPs (65%) perceived that an important skill they had developed when communicating with young people was being able to adapt their communication style to the maturity level of the young person, as well as the wishes of the parents. HCPs followed the lead of the young person to help them determine the level of detail to provide and individualize their approach to the information needs of the individual.
Being available to the families (repeat information, listen, and empower)
To build trust and rapport, 11 HCPs (55%) reported that they perceived an important skill they had developed was to patiently explain the details of the diagnosis and the treatment options and to empower the families to ask questions. HCPs reported that, given the complex nature of the disease and treatment and families’ high distress, it was necessary to patiently repeat information several times to ensure the family is understanding. Similarly, hard-to-treat cancers involve decisions about treatment options where there is a lot of uncertainty around the outcomes and decisions about whether and when to begin palliative care. HCPs reported wanting to empower the families so that they felt like they were “part of the team,” but also ensure that parents were aware that “there’s no right or wrong way in what decision is made” so they do not shoulder guilt or blame. (Social worker)
Transparency of processes: show parents they are trying everything
Four HCPs (20%) reported that it was important to show parents that they were trying everything to save their child. HCPs noted that some parents valued knowing that the healthcare team was “doing something right to the very end” (Nurse). For parents who needed more information, HCPs shared that they explained their decision-making processes to families and described any procedures that were occurring “behind-the-scenes” including, for example, molecular testing that had been “tried and hadn’t quite worked” (Oncologist), their own research into new trials, their investigations of new agents, and their consultations with colleagues overseas. HCPs reported that this was particularly important when there were long delays to starting a trial or receiving molecular results.
Research question 3: training
Ten HCPs (50%) reported that they received no specific training to deliver “bad news” to families and how to communicate with families about hard-to-treat cancers. Of the 10 HCPs (50%) who had received training, 5 were oncologists, 4 were nurses, and 1 was a social worker. These HCPs reported that training was generally undertaken via independent study (e.g., workshops and certificate of bereavement counseling) and not a formal part of their degree or specialist education. The training was also typically about breaking bad news in general.
Practical and easily accessible resources
Ten HCPs (50%) expressed a desire for training to be developed to prepare them and their colleagues for difficult discussions with families of young people with hard-to-treat cancers. HCPs emphasized that training or education resources needed to be easily accessible; “easily readable and easy to follow, not 15 pages of text” (Oncologist). Preference was given to practical training that involved role-play or simulation, or online resources that outlined topics that need to be covered when communicating with families.
Training for junior HCPs
Most HCPs reported that they developed their skills “on the job” by observing their seniors and by receiving feedback from their colleagues and supervisors. However, nurses, social workers, and oncologists with <10 years’ experience reported dissatisfaction with this approach and, more commonly than oncologists with >16 years experience, reported seeing value in the development of training resources for HCPs to assist communication with families. Oncologists with <10 years’ experience wanted training for themselves, whereas nurses and social workers discussed the importance of training both for themselves and for junior oncologists.
Discussion
We qualitatively explored HCPs’ challenges, as well as their skills and training needs for communicating with families of children with hard-to-treat cancers. HCPs commonly reported family- and systems-related challenges. Consistent with previous research, HCPs in this study revealed that they build and maintain trust and rapport with families by spending time with them, answering their questions, and repeating information (Sisk et al. Reference Sisk, Friedrich and Mozersky2020). Half the HCPs had not received any specific training for communication with families about hard-to-treat cancers, and the other half expressed a desire for the development of such training. Nurses, social workers, and less experienced oncologists appeared to be more interested in future training than more experienced oncologists.
HCPs in our study noted challenges in their communication with families, including uncertainty and balancing information provision with families’ distress and need to maintain hope. HCPs also reported that their expectations for treatment outcomes often differed to those of parents, and this was a major challenge in their interactions with families. These findings echo those from other studies, with previous research also showing that parents can have “unrealistic” expectations for treatment in poor-prognosis childhood cancer (Kaye et al. Reference Kaye, Stall and Woods2021; Mack et al. Reference Mack, Cronin and Uno2020). This suggests discrepancies in clinician–parent understanding (Sisk and Kodish Reference Sisk, Kodish, Mazur and Berg2020). Furthermore, prior research shows that parents’ understanding of poor prognosis may improve when oncologists provide direct statements about incurability, as opposed to indirect or implied prognostic language (Kaye et al. Reference Kaye, Stall and Woods2021). Honest disclosures of poor prognosis can actually help relieve uncertainty in parents, as it allows them to make fully informed decisions about their child’s treatment (Nyborn et al. Reference Nyborn, Olcese and Nickerson2016). Indeed, parents who find prognostic information to be particularly upsetting still want to receive this information (Mack et al. Reference Mack, Wolfe and Grier2006), which underscores the importance of honesty.
HCPs reported that online information about novel therapies and clinical trials overseas exacerbated the divide in expectations between themselves and parents. Previous literature shows that many parents of children with cancer (though not necessarily hard-to-treat cancers) use social media to source information about treatment options (Gage-Bouchard et al. Reference Gage-Bouchard, LaValley and Warunek2018; Gün and Şenol Reference Gün and Şenol2019; Nagelhout et al. Reference Nagelhout, Linder and Austin2018), despite the commonly inaccurate, unproven, or irrelevant nature of this information (Domínguez and Sapiña Reference Domínguez and Sapiña2015; Gage-Bouchard et al. Reference Gage-Bouchard, LaValley and Warunek2018). Given these findings, it is important that HCPs have the resources and skills to help parents navigate the varying quality of information on social media, which may then help to ensure that parents have realistic expectations for their child’s treatment outcomes.
Another major challenge reported by HCPs occurred when parents wanted to censor the information HCPs provided to the young person with cancer. Most pediatric cancer patients want to be involved in conversations about their cancer care, at least to some extent (Coyne et al. Reference Coyne, Amory and Kiernan2014), and adolescents may want to take part in the treatment decision-making process (Blazin et al. Reference Blazin, Cecchini and Habashy2018; Wakefield et al. Reference Wakefield, McLoone and Fleming2011; Weaver et al. Reference Weaver, Baker and Gattuso2015). Adolescent and young adult (AYA) cancer patients also highly value receiving honest prognostic information from their HCPs (Mack et al. Reference Mack, Fasciano and Block2018). Indeed, there are legal and ethical implications for communication with AYAs, which are beyond the scope of the current study, and are thoroughly explored in previous research (Sansom-Daly et al. Reference Sansom-Daly, Wakefield and Patterson2020; Sisk et al. Reference Sisk, Canavera and Sharma2019a). Furthermore, when parents want to censor information from their child, it conflicts with HCPs’ preference for honest and developmentally appropriate communication with young people as reported by the HCPs in this study as well as in previous studies with pediatric cancer patients (Lin et al. Reference Lin, Gutman and Hanson2020) and HCPs (Laronne et al. Reference Laronne, Granek and Wiener2022).
HCPs revealed that social workers, and at times nurses, may be missing from important conversations with families due to staff shortages and a perceived lack of understanding of the social worker role among the rest of the clinical team. This differs from previous research in the United States where pediatric oncology social workers perceived themselves to be well integrated within the treating team (Jones et al. Reference Jones, Currin-Mcculloch and Pelletier2018). Social workers are key members of the multidisciplinary pediatric oncology team and help to ensure a holistic approach to care for the child with cancer, their parents, and their siblings (Jones et al. Reference Jones, Currin-Mcculloch and Pelletier2018). Furthermore, research shows that early integration of palliative care can help families better adjust when treatments are not working (Rosenberg et al. Reference Rosenberg, Wolfe, Jones, Abrams, Muriel and Wiener2016); a point echoed by the social workers in the current study. Research shows that treating oncologists are typically responsible for deciding when to introduce palliative care to the families, but that social workers can help advocate for early integration of palliative care concepts (Dalberg et al. Reference Dalberg, Jacob‐Files and Carney2013; Jones et al. Reference Jones, Currin-Mcculloch and Pelletier2018).
Implications
HCPs’ challenges to communicate with families about hard-to-treat cancers may lead to burnout and communication that does not meet families’ needs (Sisk et al. Reference Sisk, Schulz and Kaye2022). Communication training may help to address some of these challenges (Sisk et al. Reference Sisk, Mack and DuBois2019b) if it is feasible to complete. Without systematically provided training, individual HCPs are required to be motivated to actively seek out their own education about communication and the latest research (Hopia et al. Reference Hopia, Miettinen and Miettinen2019), and lack of time can be a key barrier to doing so (Ten Bruggencate et al. Reference Ten Bruggencate, Kremer and Caron2009). This may result in inconsistent approaches to communication with families across different HCPs (Kaye et al. Reference Kaye, Stall and Woods2021).
Furthermore, while existing communication training regarding “breaking bad news” may be beneficial for HCPs (Johnson and Panagioti Reference Johnson and Panagioti2018), there are no known evidence-based communication interventions that specifically focus on the pediatric oncology setting (Sisk et al. Reference Sisk, Schulz and Mack2019c). Targeted communication training for pediatric oncology HCPs is important given the many differences between standards of care in adult medicine versus pediatric medicine (Hussen Reference Hussen2022), as well as the increasingly complex and difficult to understand treatments for hard-to-treat cancers. Limited evidence suggests that such training may be feasible (Kaye et al. Reference Kaye, Cannone and Snaman2020). Future research should develop and evaluate resources that are easy to use, support HCPs’ communication with families about hard-to-treat cancers and complex treatment schedules, facilitate interdisciplinary communication, and address the many challenges associated with such communication. Education may include strategies to improve families’ prognostic understanding such as “seed planting” whereby the poor prognosis is discussed early, and further prognostic information is gradually provided over time (Aglio et al. Reference Aglio, Woods and Baker2023; Kaye et al. Reference Kaye, Stall and Woods2021). Education may also incorporate support for HCPs regarding how to advise parents to navigate information on social media about their child’s prognosis and treatments.
Limitations
Recruitment to the study began at one site, and participants were colleagues of our clinical lead investigator, which may have skewed the results and overrepresented HCPs working at the study site. Nevertheless, we used the snowballing method to help overcome this and increase the reach of our data collection beyond the main study site. Furthermore, given the limited pool of eligible participants, there was likely to be minimal sampling bias. Only 3 social workers participated in the interviews. Given the important perspectives regarding multidisciplinary teamwork reported by the social workers in the current study, future research should actively engage social worker perspectives. In addition, our study did not include psychologists, palliative care physicians, child-life therapists, and several other allied HCPs. It is likely that the entire multidisciplinary team would benefit from communication skills training, as emotional support for families (using associated communication skills) could be expected to be provided by any member of the team.
Conclusion
With a growing number of complex treatment options for children’s hard-to-treat cancer and easy access to potentially inaccurate or irrelevant treatment information online, resources are needed to support HCPs to communicate with families. Such communication resources may be particularly beneficial for less experienced oncologists during their training and should aim to prepare them for the typical challenges they will face and the skills they will need, as well as foster greater collaboration with nurses and social workers. This will ensure equitable access for children and families to the best standard of communication across their treatment trajectory.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S1478951523001992.
Data availability statement
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by Luminesce Alliance – Innovation for Children’s Health for its contribution and support. Luminesce Alliance is a not-for-profit cooperative joint venture between the Sydney Children’s Hospitals Network, the Children’s Medical Research Institute, the Children’s Cancer Institute, the University of Sydney, and the University of New South Wales Sydney. It has been established with the support of the NSW Government to coordinate and integrate pediatric research. This study was funded by a Translational Program Grant from the Cancer Institute NSW. Prof. Wakefield is supported by the National Health and Medical Research Council of Australia (APP1143767).
Competing interests
The authors declare that they have no conflict of interest.
