Specialist cancer services for teenagers and young adults in England: BRIGHTLIGHT research programme

Provision of teenage and young adult cancer care in England

Specialised teenage and young adult (TYA) cancer services in the UK have evolved over the past 30 years, with much input from the charity Teenage Cancer Trust. 20,21 Since the 1960s, a model of delivering cancer care to children has been established,11 and in 2001 adult cancer services were reconfigured into cancer networks, resulting in improvements in both patient experience and outcomes. 22 The first teenage cancer unit opened in the Middlesex Hospital in 1990, but it was not until 2005 that national policy was published that formalised the configuration of services specifically for teenagers and young adults. 23

The release of the National Institute for Health and Care Excellence (NICE) improving outcomes guidance (IOG) for children and young people with cancer in 200523 was a significant landmark on the landscape of English TYA cancer service development. This outlined detail about the provision of services for teenagers and young adults, such as clinical organisation, facilities, and diagnostic and therapeutic modalities. 23 Although this guidance supported subsequent service delivery, the evidence review that underpinned it was a collation of evidence on child and adult cancer services, some of which was assessed to be of fair to poor quality. Of the 15 pieces of evidence reviewed, only two were specific to teenagers and young adults. 24

Despite the evolution and development of UK cancer services for young people, there continued to be variation regarding where young people with cancer received their care. 25 Even with government recommendations advocating ‘young person-friendly’ health services,26 many young people in the UK were cared for on adult wards27 or in children’s services28 (Figure 2). It was suggested that it was ‘inappropriate’ to deliver care to young people in either child or adult environments of care,30,31 or in settings not equipped to meet their needs. 32

FIGURE 2.

An illustration of the three types of health service where young people with cancer may be cared for. Reproduced with permission from Sarah Lea. 29 This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) license, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: https://creativecommons.org/licenses/by/4.0/. The figure below includes minor additions and formatting changes to the original figure. Images reproduced from Nottingham University Hospitals NHS Trust, 2017 with permission (Nottingham University Hospitals NHS Trust, 2021, personal communication).

Reproduced with permission from Sarah Lea. 29 This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) license, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: https://creativecommons.org/licenses/by/4.0/. The figure below includes minor additions and formatting changes to the original figure. Images reproduced from Nottingham University Hospitals NHS Trust, 2017, with permission (Nottingham University Hospitals NHS Trust, 2021, personal communication).

Prior to the release and implementation of the IOG, approximately 52% of young people in England had been receiving care in a hospital that became a principal treatment centre (PTC). This group mainly comprised teenagers (aged 15–18 years) rather than young adults (aged 19–24 years). The place of care was a key focus within the IOG,23 which stated young people should be treated in an ‘age-appropriate environment’ and have access to ‘age-appropriate’ facilities. What made an environment or facility age-appropriate was not defined. However, specialist services could not be mandated without sufficient evidence to underpin them. This led to the process of designation (i.e. cancer services for young people in England were structured into networks of care with a central TYA-PTC as a ‘hub’ of expertise and hospitals with adult cancer services surrounding the TYA-PTC could apply to be ‘designated’ to provide cancer care to young people aged 19–24 years).

The model of service delivery for young people with cancer in England in 2012, when this programme of work began, consisted of 13 TYA networks of care with a TYA-PTC and varying numbers of associated designated hospitals (Figure 3). The TYA-PTC provided treatment expertise across the range of cancers common in young people, supported by a dedicated TYA multidisciplinary team (MDT) to meet the psychosocial needs of this population, within an environment that was tailored to the developmental and social needs of young people. 21 Young people aged up to 16 years could receive care in a children’s PTC or a paediatric oncology shared care unit that was authorised to provide certain aspects of supportive care, such as administration of blood products or simple chemotherapy drugs. Young people aged 19–24 years were to have ‘unhindered access’ to age-appropriate care and had the choice of being referred to a TYA-PTC or to stay in an adult cancer unit in a designated hospital within the network (Figure 4).

FIGURE 3.

Map of the location of the 13 TYA cancer networks in England. (1) Cambridge, (2) Bristol, (3) Oxford, (4) Liverpool, (5) Newcastle, (6) East Midlands, (7) Birmingham, (8) Southampton, (9) Leeds, (10) Manchester, (11) South Thames, (12) North Thames and (13) Sheffield.

FIGURE 4.

The range of places where a young person with cancer may receive their care, dependent on whether they are aged 13–18 or 19–24 years. Young people may receive access to specialist TYA cancer care (highlighted by the solid arrows), but young people may still be cared for in either child or adult cancer services (highlighted by the dashed arrows). Reproduced with permission from Sarah Lea. 29 This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) license, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: https://creativecommons.org/licenses/by/4.0/. The figure below includes minor additions and formatting changes to the original figure.

Reproduced with permission from Sarah Lea. 29 This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) license, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: https://creativecommons.org/licenses/by/4.0/. The figure below includes minor additions and formatting changes to the original figure.

There was a requirement that designated hospitals would notify the TYA-PTC of young people newly diagnosed with cancer so that there was ‘sharing of responsibility for patient management’ between the tumour site-specific clinical team at the designated hospital and the experts at the TYA-PTC. 33 Moreover, young people at designated hospitals should have ‘unhindered access’ to the support of the wider MDT via outreach work performed by the specialist professionals from the TYA-PTC (e.g. young people’s social workers). 23 Within each network there were also hospitals that were not allocated to provide care to TYA (non-designated hospitals). A proportion of teenagers and young adults continued to be cared for in these non-designated hospitals without access to the age-specific expertise of the TYA MDT at the PTC or access to age-appropriate care throughout their entire cancer journey. 34

Although there was a variety of services in which a young person may receive their cancer care, dependent on their disease, age, location and availability of services, there was variation in how this was translated and implemented across England. Of the 76 hospitals that were designated to deliver TYA cancer care, in 2013 approximately one-third were unable to deliver ≥ 50% of the standards that had been specified for being designated. 35 There were no consequences of this, and these hospitals have remained designated for teenagers and young adults, despite lacking many elements of a young adult-friendly cancer service. 36

BRIGHTLIGHT

The guidance and policy directing TYA cancer services in England in 2010 was not based on evidence and there was a lack of research evaluating how well these TYA cancer care networks operated. BRIGHTLIGHT is the applied health research programme providing this evaluation, which evolved from feasibility work in the Essence of Care study in 2009/10 that informed the methods for BRIGHTLIGHT. 37–41 During the Essence of Care study, we identified wide variation in the delivery of care across England, which informed the need for all the studies within the programme to be multicentre as well as containing longitudinal aspects. Although we have achieved this, we have been challenged by recruitment to the cohort and changes in regulatory processes throughout the study period, resulting in delays and preventing the detailed analysis of the cohort data necessary to fully understand the results. As a result, we provide a tentative conclusion in the knowledge that further work may be required to further inform answers to the overarching question: do specialist cancer services for teenagers and young adults add value?

Objectives

The programme of research was divided into three workstreams (Figure 5).

FIGURE 5.

Flow diagram depicting the inter-relationships between the workstreams. HES, Hospital Episode Statistics.

Workstream 1: establishes the description of teenage and young adult cancer care including specialist care

1. 1.1. Explore the culture of care through non-participant observation, semistructured interviews and analysis of departmental documents.

2. 1.2. Identify the specialist competencies and added value of specialist health professionals through an international Delphi survey.

3. 1.3. Develop the TYA Cancer Specialism Scale to categorise three levels of TYA care and apply to individual patient-level data.

Workstream 2: examines care of young people with cancer in a cohort study

1. 2.1. Relate the level of cancer care received by teenagers and young adults to quality of life (QoL), satisfaction with care, clinical processes and clinical outcomes (overall, by age group and by tumour type).

2. 2.2. Examine young people’s experience of cancer care through a longitudinal descriptive survey.

3. 2.3. Compare social and educational milestones among young people receiving different levels of TYA cancer care.

4. 2.4. Examine geographic and sociodemographic inequalities in access to TYA cancer care.

Workstream 3: examines the economics of the levels of teenage and young adult cancer care

1. 3.1. Calculate detailed costs to the NHS and Personal Social Services of teenagers and young adults receiving different levels of cancer care.

2. 3.2. Estimate the cost incurred by teenagers and young adults and families receiving different levels of cancer care.

3. 3.3. Calculate the cost-effectiveness of different levels of cancer care.

Young people

Young people working as co-researchers during the feasibility work lent itself well to this short intensive piece of work. However, we felt that the sustainability of this model was not viable given the life-stage commitments of young people and the length of the proposed programme. Our application stated:

An initial workshop will be held with the Young Persons Reference Group; thereafter they will be consulted using options such as email discussion rather than face-to-face meetings, recognising the range of life stage commitments of TYA [teenagers and young adults].

Discussions with young people between December 2011 and August 2012 revealed that e-mail would not be an appropriate or responsive method of involvement, as e-mail was no longer ‘vogue’ (YAP workshop participant) and had been replaced by Facebook. Young people also advocated some face-to-face contact to foster relationships between themselves and the research team. We opted for a closed Facebook page, which took 14 months to be approved by our NHS trust and opened in October 2013, 13 months after the 2012 workshop. This delay was unfortunate, as it meant that many young people who attended the 2011/12 groups had moved on/changed contact details in this time.

We accessed young people through the conference ‘Find Your Sense of Tumour (FYSOT)’. 43 Annually between 2008 and 2017, we consulted around 200 young people aged 13–24 years on study design, suitability of research questions, outcome, recruitment approach and dissemination. 42,44,45 Young people gave their opinions individually and anonymously through handheld device surveys. We also recruited four young people from FYSOT to perform in our dissemination event ‘There is a Light’.

Parents and siblings

We planned to include parents and siblings and, following notification of funding, we visited existing groups of young people (Leeds/Birmingham) and family events (Cambridge). The Cambridge event involved focus groups with young people and parents/siblings. In our original plan, we stated our intention to report back to parents annually. In response to recruitment problems we focused efforts on understanding the barriers to recruitment from young people and professionals. Additionally, we added a further strand of data collection in response to our PPI work to more fully understand the experiences of carers.

We have involved around 1200 people in our PPI strategy (not including professionals) (Figure 6).

FIGURE 6.

A schematic diagram of BRIGHTLIGHT PPI involvement.

Networks

The care of young people is complex with low incidence rates, multiple cancer types and care delivered across children, TYA and adult services in over 100 NHS trusts; consequently, project delivery was always going to be challenging. In anticipation of this we developed a third PPI group of ‘networks’. These included clinical/non-clinical staff delivering care, research network staff, adult PPI groups and charitable organisations.

Facebook

The YAP remains connected through the closed Facebook page, which was monitored daily by the lead for PPI or a delegated team member. Contact was predominantly via group messenger to advertise workshops, post questions and advertise relevant activities and opportunities to participate in other projects. On only one occasion did the lead for PPI have to intervene because of inappropriate postings.

Workshops

Workshops have evolved in response to feedback from young people. 46 Changes included sending out the full aims, objectives and agenda prior to the workshop, allowing more time for young people to network, trying to engage a wider group of young people, varying the times, days and location of workshops and playing quiet music during activities and breaks. The typical workshop structure can be seen in Appendix 1:

The staff were amazing, very informative. Made me feel included throughout. It was a relaxed informal atmosphere which didn’t make the goal of the day seem to [sic] strenuous or daunting. Also thanks for the voucher MERRY CHRISTMAS [drawing of love heart].

The research team were actively involved in workshops, taking part in role play to illustrate some of the issues around recruitment. We piloted this in 2013 as an alternative to updating young people on the progress of the study through a presentation using PowerPoint^® (Microsoft Corporation, Redmond, WA, USA), and it received favourable feedback. In 2014, we re-enacted a meeting held with the NIHR to illustrate further points about recruitment/retention. Workshops were followed up within 48 hours with an e-mail from the PPI lead to enquire about any distress/upset. No distress was reported, with young people reporting that they felt that the workshops were therapeutic and enjoyed the opportunity to contribute to the research agenda and meet with other young people.

Review of documents and development of survey materials

The YAP contributed to survey development and approach materials through focus groups and telephone cognitive interviewing. 47 During the set-up of study 4, comments from young people included amending the approach period, wording on the patient information leaflet and formatting of the consent forms allowing young people to tick boxes rather than initialling, and we have now introduced this to all our consent forms. Parents also commented on survey materials during set-up and appeared more sensitive to questions than young people. Questions that parents felt were insensitive were double-checked with young people who had no problem with them.

Branding

The YAP were responsible for rebranding the study from the ‘2012 TYA Cancer Cohort Study’ to BRIGHTLIGHT in response to young people noting that the study name needed to be memorable. 44

Interventions for recruitment

Recruitment to BRIGHTLIGHT began in July 2012, with early signs of lower than anticipated recruitment. Despite this, our acceptance rate by patients approached to participate was 80% and we believe this high level of acceptability was due to our extensive PPI strategy during study design and set-up. We turned to our PPI partners for recruitment advice. Young people suggested a more appealing patient information sheet, audiovisual information, advocating informed choice by publicising that professionals should facilitate young people’s awareness of all research studies available to them and recruitment by other members of the treatment team. In response to these suggestions, we designed a shorter patient information sheet, which was given out with the paper document, videos of the information sheet and a ‘meet the team’ section on the website. Recognising that the YAP are an engaged group, we sought validation of suggestions around approach and recruitment with participants of FYSOT in 2013. 45 We also trained a group of social workers/youth support workers to gain consent. Lastly, we held a workshop with youth support co-ordinators and enrolled our networkers to contribute to a weekly Twitter (Twitter, Inc., San Francisco, CA, USA; www.twitter.com) recruitment campaign. However, this was not as successful as Twitter recruitment had proved for research on other diseases. 48–50

Interventions for retention

Uptake to participate in the BRIGHTLIGHT cohort among those approached was higher than anticipated, as was retention at wave 2 (i.e. 12 months after diagnosis). However, we noted falling retention rates at wave 3 (i.e. 18 months after diagnosis) to around 30%. In response, we focused our 2014 workshop on strategies to improve retention and introduced interventions including personalised letters to participants and feedback of results using info graphics. After implementation of these measures, retention rates increased to 60%. 51

Dissemination and publicity

The YAP have contributed to our dissemination and publicity programme, including presentation of emerging results by two YAP members at FYSOT (link to FYSOT presentation52); chairing the BRIGHTLIGHT/TYAC (Teenage and Young Adult with Cancer professional organisation) conference in July 2017; and an interview on BBC Radio 5 Live (BBC, London, UK). National dissemination by a novel route was achieved through collaboration with Dr Brian Lobel, Contact Young Company and four young people with a previous cancer diagnosis. 53 The collaboration involved a series of workshops with the research team, young people and theatre group to create an artistic interpretation of emerging BRIGHTLIGHT results. Additional funding was secured from the Wellcome Trust and Macmillan Cancer Support, which allowed the performance to tour professional and patient conferences, including FYSOT in 2017. The performance received great reviews in terms of content, artistic talent and accessibility of results. 54 The young people participating also expressed benefit from being part of the team. 55

New studies

The YAP have contributed to new study development, which addressed online information needs,56 end-of-treatment concerns57 and the development of a sarcoma-specific patient-reported outcome measure. 58,59 Additionally, the YAP commented that the impact of cancer on sex, body image and relationships was poorly addressed; thus, we held two workshops with a view to new study development. 60 This has resulted in a patient-initiated study examining sexual health in young people with cancer, which was unsuccessful in the first grant submission but further applications for funding are planned. YAP members who wish to pursue a research career are co-applicants on new studies.

We do not have any negative effects to report from our PPI strategy. However, we do recognise the commitment and resource required from the research team to ensure that young people are included effectively in the study.

Diversity

The gender balance of the YAP was predominantly female, a phenomenon not unique to BRIGHTLIGHT. Although we had some ethnic variation within the group (around 20% of the YAP were ethnic minorities), it did not always reflect this diversity due to the variability of workshop attendance. Attendance at workshops was generally in the region of 9–10 young people, which was around half of our YAP. We sought feedback and were given multiple reasons including ‘too rainy to attend’, ‘hottest day of year, too hot to attend’, ‘would be better in summer holidays’, ‘would be better if not in summer holidays’. After several attempts of varying the times, calendar month and location, we felt that there was no real preference and workshops are now held on a Friday/Saturday in London (all expenses are paid and a voucher is given for attendance).

Bereavement

There is always a risk when bringing young people with cancer together over time that bereavements may occur within the group. Three deaths occurred within our group. We arranged for the local Macmillan Support and Information Service to offer bereavement support; however, this was not taken up (to our knowledge) and because of patient confidentiality, we would not be told of any young people who had approached the service.

Unanticipated events

Several unexpected events occurred. First, the lengthy and bureaucratic NHS processes meant that some of the suggestions that young people wanted could not be implemented or took considerable amounts of time: for example, shortening the patient information sheets, implementation of the Facebook page and utilising social workers/youth support co-ordinators for recruitment. Meeting expectations of the YAP, particularly around the use of emerging technology and social media, was constrained by resources and internal governance issues. After the Facebook page was established, young people requested Twitter, then Instagram (Facebook, Inc., Menlo Park, CA, USA; www.instagram.com), then Snapchat (Snap Inc., Santa Monica, CA; www.snapchat.com), curation of which would have required a dedicated communications team. Therefore, we did not implement a Snapchat account and our Instagram account did not really flourish.

Following our feasibility work, policies were put in place for codes of conduct (e.g. alcohol, smoking, drug use) and expenses. Further to this, several policies were implemented in response to events. An ‘out-of-hours’ policy was introduced in response to a young person having to make an emergency trip home late at night following sickness of her child during a 2-day workshop. Second, a ‘sickness policy’ was implemented after a young person (> 16 years) was admitted to hospital via accident and emergency (A&E) during a workshop and refused to give a next of kin contact. Finally, although no distress has been reported after the workshops, in response to an e-mail follow-up after a workshop, we had one incident of significant distress during dinner after a workshop. This young person was referred to their treatment team for support.

Round 1 questionnaire

A comprehensive list of competencies was generated from our preliminary study,38 and this formed the content for the first round of this Delphi survey. These were subdivided into skills, knowledge, attitudes and communication. All the questions had closed-ended responses using 9-point Likert scales (i.e. strongly agree to strongly disagree). However, as the competency list was initially generated by health-care professionals based in the UK, several open-ended questions were included to ensure that the survey would accommodate the opinions of professionals in other countries. The questionnaire was administered through a web-based survey programme.

Round 2 questionnaire

Only items for which there was no agreement were included in the round 2 questionnaire. Qualitative content analysis was used to analyse responses to open-ended questions. Panel members were also requested to identify five skills, areas of knowledge, and attitudes that they considered the most important.

Data analysis

Items were ranked and reported according to medians. Medians of 7–9 were defined as strong support, 4–6.5 as moderate support and 1–3.5 as weak support. The mean absolute deviation from the median was calculated and the level of agreement was categorised according to thirds of the mean absolute deviation from the median (low > 1.41, moderate 1.08–1.41, high < 1.08). These summaries were also calculated according to profession (medical doctor, nurse, other health-care professional) and differences determined using chi-squared tests to compare the number of respondents who had strong agreement (consensus).

The physical and social environments of care

The décor, structure, function and facilities of a physical environment tailored specifically to the needs of young people were highlighted as important. Health-care professionals and teenagers and young adults agreed on similar aesthetic features, for example colourful décor that created a less clinical atmosphere. Professionals described how wards in the TYA-PTC provided modern, colourful and well-resourced environments, and these settings facilitated health-care professionals to enjoy interactions with their patients and colleagues. Health-care professionals identified that the physical surroundings in which they interacted with their patients had an impact on the conversations that they had. Young people described the atmosphere on these wards as calm, relaxed and homely, which promoted normality.

There was a relationship between the physical environments in which young people were cared for, their experiences of care and the social relationships that they built with those around them. The dedicated social spaces provided in the TYA-PTC enabled youth support co-ordinators to better fulfil their role, where the environment provided the space and facilities to bring young people together.

Communication and core values

Communication was the major visible process of care that occurred between young people and professionals, both at an individual level and within groups. Three types of communication emerged:

1. Interpersonal – young people recognised effective communication, and they identified and valued the relationships they built with health-care professionals. Young people recognised and valued the meaningful interactions that they had with health-care professionals, both within TYA-PTCs and designated hospitals. Young people described continuity and consistency with health-care professionals as advantageous, enabling them to build relationships with those caring for them. Likewise, health-care professionals acknowledged that continuity of care was important. Youth support co-ordinators described an essential part of their role was being a ‘constant’ for patients. Continuity of staff and the implementation of routine communication processes, such as effective handovers between professionals, meetings and discussion groups provided opportunities for a united and knowledgeable health-care team to form and to flourish.

2. Intra-hospital – there were multiple and separate circles of intraprofessional communication about care within a hospital, involving all members of the direct care team and others. Processes described by health-care professionals were more complex where multiple teams were involved. Young people spoke specifically about the role of the clinical nurse specialist and their part in ensuring care was not disjointed.

3. Hospital to hospital – communication was back and forth between the teams at the designated hospitals, shared care hospitals and the TYA-PTC, ensuring that both the young person’s clinical and their psychosocial needs were being met. There was variation in processes that affected overall experience.

Three core values emerged: recognising individuality, promoting normality and empowering young people. These core values were an essential part of the less visible ‘below the surface’ culture, and were values that underpinned TYA cancer care across all settings. Delivery of care tailored to the individual patient’s needs and effective provision of information were highlighted as important by young people and professionals. Health-care professionals sought to encourage those young people who were cognitively able to have some control over their care and the decisions made about it.

The development of health-care professional holistic competence and the culture of care

The formation and sharing of a culture where care was responsive to the unique needs of the teenagers and young adults was influenced by four factors: a consistent volume of young people using services, effective leadership, an appropriate and accepting attitude, and patience.

Consistency and a large number of young people using a service was important in the formation of an age-appropriate, young person-centred culture of care. The TYA-PTCs hosted a consistent, concentrated volume of young people, compared with many of the children’s and adult cancer settings. In all contexts, leadership was essential to shape and perpetuate the culture of care. Leaders were vital in bringing together the whole team, creating a culture in which all health-care professionals communicated with each other effectively. Leaders were important in assisting the formation of trusting relationships between all members of the team. Shared beliefs and ‘buy-in’ of health-care professionals into what was different and special about caring for young people with cancer were core to the culture of care. It took considerable time for such connections and knowledge about caring for young people to develop on both a network and a local level, particularly as these networks could span a variety of specialties: adult, child and a wide range of tumour site-specific teams.

The importance of the core values that underpin care, and the need for education, effective leadership and multidisciplinary teamworking was described as essential. These should be prioritised when developing and evaluating interventions that contribute to the delivery of care. Care delivered in an environment that promotes normality through facilitating socialisation with peers was described as essential to the delivery of optimal holistic and young person-centred care. Growing and nurturing a culture of care that meets the unique needs of young people with cancer and improves their experiences of care takes time and commitment.

Research methods

Over the period of the programme grant we have learnt much about conducting research involving young people with cancer. The evaluation from the perspective of young people utilised a longitudinal cohort design on the basis that it was not possible to manipulate where young people received care but we needed to capture how the care received at diagnosis influenced perceived outcomes in the short and medium terms. Although this afforded us the opportunity to gather a wealth of valuable data, this has taken many years to be completed. For this reason, we were unable to capture the influence of the dynamic changes in services. The complexities of researching health care have become recognised in recent years with the emergence of methods specifically designed to capture this, such as rapid ethnography93 and applied systems thinking. 94 We would therefore recommend using these or similar methods to capture the complexities of the relationships between stakeholders and the environment.

We have published extensively on involving young people in research, the barriers to this and the potential mechanisms to overcome these barriers. 45,66,67 However, we found that, even with proactive changes to the protocol, these mechanisms did not always have the desired benefit of improving recruitment; yet the rate of refusal of young people to join the cohort was low,63 and our experience with the NIHR TYA stakeholder meeting and National Cancer Research Institute (NCRI) TYA research groups has been that there is much enthusiasm from young people for supporting initiatives to increase TYA opportunities to have access to research. It may be that a single change in research process is not sufficient to affect improvements in study accrual, and a multipronged approach is required. We would recommend that this is guided by a conceptual framework, such as the 5As model proposed by Fern et al. 76 The 5As are availability (cancer trials are not always available for teenagers and young adults because of the rare cancer types they present with), appropriateness (studies often have inappropriate age eligibility criteria applied, which is based on whether the study has originated from children’s or adult oncology/haematology), accessibility (studies are often not open where young people are being treated), awareness (increased awareness among health-care professionals and teenagers and young adults about the importance of trial entry is required) and acceptability (the study question and trial design must be acceptable to both the professional and young person). 76 These 5As can help to identify where the barriers lie and guide where interventions and changes in service delivery may be of benefit.

Research arising from the programme

A priority for further research has emerged that links results from workstreams 1 and 2 and is immediately relevant to the definition of TYA services currently envisaged in draft NHS England service specifications for both TYA-PTCs and designated hospitals. A study that reproduces the methodology used by BRIGHTLIGHT but updated with data from TYA diagnosed in 2021 would determine whether or not the results seen at the first wave of data collection in the cohort are being reproduced with a similar differential of outcomes between NONE, SOME and ALL groups. If this was so, there would be a strong argument for urgent exploration of those poorly defined factors that determine ‘membership’ of these groups. Alternatively, there may no longer be separation between these groups, a result that might be consequent on an improved access to the key elements of specialist care as described here, including outside the TYA-PTC. Simultaneously, rapid evaluation of cancer services for teenagers and young adults should be conducted to determine whether or not the culture of TYA care has, in fact, evolved. Through the use of rapid methods it would be possible to explore interactions and co-ordination of care in most, if not all, TYA-PTCs. This would capture the variation in service delivery and the potential to link patient-reported outcome and experience to the care they receive.

We have reported in detail the limitations of the TYA Cancer Specialism Scale and how we defined the categories of care. 68 This could be refined through including outpatient data, not just APC data, and categorising ‘specialist’ based on the hospital where young people received care, not just the trust. The SOME group warrants further investigation to determine if there is an optimum proportion of care in a TYA-PTC that has benefit rather than the crude classification of 0%, 100% and 1–99%.

We defined ‘specialist’ as care delivered in a TYA-PTC; however, these are all different and there are non-TYA-PTCs that have more TYA-specific facilities than a TYA-PTC does. As we now have a definition of age-appropriate care, we are in a unique position to be able to use survey data from the cohort to ‘test’ the conceptual model. This will enable us to objectively identify which of the components of the model define age-appropriate care and which of these are associated with better outcomes. This would guide commissioners and hospital administrators/managers on what aspects of care to invest in, that would most likely be beneficial to young people.

Although we showed that the NONE-TYA-PTC group cost the least to the NHS, this cost was calculated based on the assumption that the NHS was incurring all the cost. However, much of the infrastructure of the TYA-PTCs and some of the designated hospitals is funded through the third sector (Teenage Cancer Trust 2019 annual report notes that they supported 28 specialist units across the UK). 95 A total of 122 members of the TYA health-care team in the UK are also funded through third-sector funding: nurses (n = 76), youth workers (n = 36) and MDT co-ordinators (n = 10). These are costs that would have been included in the ALL and SOME groups in this analysis. Taking these into consideration would enable the actual cost to the NHS to be calculated.

Finally, data collection from the cohort continued for 3 years; at this time we saw that QoL was comparable in all three groups, with those receiving ALL-TYA-PTC and SOME-TYA-PTC care having an improvement in QoL more rapidly to being a mean of 1–5 points different from the NONE-TYA-PTC group. What would be important to establish is whether the rate of improvement continued, in which case by 5–10 years after diagnosis, QoL could potentially be superior for the ALL and SOME groups in comparison with the NONE group. The philosophy of TYA cancer care is that support is provided to enable young people to continue their lives to achieve the milestones that they would have had they not had cancer. Three years may not have been sufficient time to be able to capture this benefit. A follow-up survey when the cohort is 10 years post diagnosis would enable this benefit to be determined. This would also afford the opportunity to explore outcomes other than QoL and survival that could have benefited from specialist care.

Peer-reviewed publications

Fern LA, Taylor RM, Barber JA, Alvarez-Galvez J, Feltbower R, Lea S, et al. Processes of care and survival associated with treatment in specialist teenage and young adult cancer centres: results from the BRIGHTLIGHT cohort study. BMJ Open 2021;11:e044854. https://doi.org/10.1136/bmjopen-2020-044854

Fern L, Taylor RM, Millington H, Ashton J, Grew T, Brooman K, et al. The art of age-appropriate care: reflecting on a conceptual model of the cancer experience for teenagers and young adults. Cancer Nurs 2013,36:E27–38. https://doi.org/10.1097/NCC.0b013e318288d3ce

Herbert A, Lyratzopolous G, Whelan J, Taylor RM, Barber J, Gibson F, Fern LA. Diagnostic timeliness in teenagers and young adults with cancer: cross-sectional findings from the BRIGHTLIGHT Cohort. Lancet Child Adolesc Health 2018;2:180–90. https://doi.org/10.1016/S2352-4642(18)30004-X

Kenten C, Martins A, Fern LA, Gibson F, Lea S, Ngwenya N, et al. Qualitative study to understand the barriers to recruiting young people with cancer to BRIGHTLIGHT: a national cohort study in England. BMJ Open 2017;7:e018291. https://doi.org/10.1136/bmjopen-2017-018291

Koo MM, Lyratzopoulos G, Herbert A, Abel GA, Taylor RM, Barber J, et al. Association of self-reported presenting symptoms with timeliness of help-seeking among adolescents and young adults with cancer in the BRIGHTLIGHT study. JAMA Netw Open 2020;3:e2015437. https://doi.org/10.1001/jamanetworkopen.2020.15437

Lea S, Gibson F, Taylor RM. ‘Holistic competence’: how is it developed and shared by nurses caring for adolescents and young adults with cancer? [published online ahead of print.] J Adolesc Young Adult Oncol 2021. https://doi.org/10.1089/jayao.2020.0120

Lea S, Gibson F, Taylor RM. The culture of young people’s cancer care in the UK: a narrative review and synthesis of the literature. Eur J Cancer Care 2019;28:e13099. https://doi.org/10.1111/ecc.13099

Lea S, Taylor RM, Martins A, Fern LA, Whelan JS, Gibson F. Conceptualizing age-appropriate care for teenagers and young adults with cancer: a qualitative mixed-methods study. Adolesc Health Med Therapeutics 2018;9:149–66. https://doi.org/10.2147/AHMT.S182176

Lea S. Perspectives: a newly qualified nurse’s experience of starting a clinical academic career. J Res Nurs 2015;20:252–5. https://doi.org/10.1177/1744987115580896

Martins A, Alvarez-Galvez J, Fern LA, Vindrola C, Gibson F, Whelan JS, Taylor RM. The BRIGHTLIGHT national survey of the impact of specialist teenage and young adult cancer care on caregivers’ information and support needs. Cancer Nurs 2021;44:235–43. https://doi.org/10.1097/NCC.0000000000000771

Stark D, Fern LA, Gibson F, Hawkins M, Hough R, McCabe MG, Taylor R. Transitioning adolescent and young adult cancer care research out of its adolescence. Eur J Cancer Care 2018;27:e12962. https://doi.org/10.1111/ecc.12962

Taylor RM, Fern LA, Barber JA, Alvarez-Galvez J, Feltbower R, Lea S, et al. Specialist age-appropriate care and quality of life outcomes in a longitudinal cohort of teenagers and young adults: the BRIGHTLIGHT study. BMJ Open 2020;10:e038471. https://doi.org/10.1136/bmjopen-2020-038471

Taylor RM, Lobel B, Thompson K, Onshila A, Croasdale M, Whelan JS, Fern LA. BRIGHTLIGHT researchers as ‘dramaturgs’: creating There is a Light from complex research data. Res Involv Engagem 2020;6:48. https://doi.org/10.1186/s40900-020-00222-5

Taylor RM, Fern LA, Barber JA, Alvarez Galvez J, Stark DP, Feltbower R, et al. Description of the BRIGHTLIGHT Cohort: the evaluation of teenagers and young adult cancer services in England. BMJ Open 2019;9:e027797. https://doi.org/10.1136/bmjopen-2018-027797

Taylor RM, Aslam N, Lea S, Whelan JS, Fern LA. Optimizing a retention strategy with young people for BRIGHTLIGHT, a longitudinal cohort study examining the value of specialist cancer care for young people. J Adolesc Young Adult Oncol 2017;6:459–69. https://doi.org/10.1089/jayao.2016.0085

Taylor RM, Fern LA, Aslam N, Whelan JS. Direct access to potential research participants for a cohort study using a confidentiality waiver included in UK National Health Service legal statutes. BMJ Open 2016;6:e011847. https://doi.org/10.1136/bmjopen-2016-011847

Taylor RM, Solanki A, Aslam N, Whelan JS, Fern LA. A participatory study of teenagers and young adults’ views on access and participation in cancer research. Eur J Oncol Nurs 2016;20:156–64. https://doi.org/10.1016/j.ejon.2015.07.007

Taylor RM, Feltbower RG, Aslam N, Raine R, Whelan JS, Gibson F. Modified international e-Delphi survey to define healthcare professional competencies for working with teenagers and young adults with cancer. BMJ Open 2016;6:e011361. https://doi.org/10.1136/bmjopen-2016-011361

Taylor RM, Fern L, Solanki A, Hooker L, Carluccio A, Pye J, et al. Development and validation of the BRIGHTLIGHT Survey, a patient-reported experience measure for young people with cancer. Health Qual Life Outcomes 2015;13:107. https://doi.org/10.1186/s12955-015-0312-7

Taylor RM, Pearce S, Gibson F, Fern L, Whelan J. Developing a conceptual model of teenage and young adult experiences of cancer through meta-synthesis. Int J Nurs Stud 2013;50:832–46. https://doi.org/10.1016/j.ijnurstu.2012.09.011

Vindrola-Padros C, Taylor RM, Lea S, Hooker L, Pearce S, Whelan J, Gibson F. Mapping adolescent cancer services: how do young people, their families, and staff describe specialized cancer care in England? Cancer Nurs 2016;39:358–66. https://doi.org/10.1097/NCC.0000000000000315

Presentations linked to BRIGHTLIGHT outside the programme objectives

Fern LA, Taylor RM, Gibson F, Millington H, Whelan J. ‘It’s All in the Name’, Branding a Longitudinal Cohort Study with Young People: BRIGHTLIGHT (Poster Presentation). Liverpool: NCIN Annual Conference; 4–7 November 2012.

Fern LA, Taylor RM, Gibson F, Millington H, Whelan J. ‘It’s All in the Name’, Branding a Longitudinal Cohort Study with Young People: BRIGHTLIGHT (Poster Presentation). Nottingham: TYAC Spring Education Day; 20 January 2012.

Koo M, Lyratzopoulos G, Herbert A, Taylor R, Barber J, Gibson F, et al. Self-Reported Presenting Symptoms and Timeliness of Help-Seeking Among Teenagers and Young Adults with Cancer: Findings from the BRIGHTLIGHT Cohort. 51st Lyon: Congress of the International Society of Paediatric Oncology (SIOP); October 2019. Pediatr Blood Cancer 66:V105 SIOP19-1019.

Koo M, Lyratzopolous G, Taylor R, Herbet A, Barber J, Gibson F, et al. Presenting Cancer Symptoms and Diagnostic Timeliness Among Teenagers and Young Adults: Findings from the BRIGHTLIGHT Cohort (Poster). Sydney, NSW: 3rd Global AYA Cancer Congress; 4–6 December 2018.

Lea S, Taylor R, Gibson F. Starting Out on a Clinical-academic Pathway: The Experience of a Newly Qualified Nurse. Abstract G227 (Poster Presentation). Birmingham: Royal College of Paediatrics and Child Health Annual Conference; 2015.

Lea S, Taylor R, Gibson F. G227(P) Starting out on a clinical academic pathway: the experience of a newly-qualified nurse. Arch Dis Childhood 100(Suppl. 3):A96.

Martins A, Taylor R, Lobel B, McCann B, Soanes L, Whelan J, Fern L. ‘Does that Mean that Cancer Patients Can’t Have Sex?’ – Adolescent and Young Adults’ Support and Information Preferences (Poster) . Sydney, NSW: 3rd Global AYA Cancer Congress; 4–6 December 2018.

Ngwenya N, Kenten C, Jones L, Gibson F, Pearce S, Stirling C, et al. ‘Hospice Scares the Life Out of Me’: Breaking Down Barriers to Communication with Young Adults with Terminal Cancer (Oral Presentation). Montreal: 21st International Congress on Palliative Care; 18–21 October 2016.

Solanki A, Taylor R, Aslam N, Gibson F, Whelan J, Fern L. The Challenges and Reality of Patient and Public Involvement (PPI) with Teenagers and Young Adults (Poster Presentation). Nottingham: 2015 RCN International Research Society Conference; 21st April 2015.

Taylor RM, Lobel B, Thompson K, Whelan JS, Fern LA. Working with Young People to Creatively Disseminate a National PROMS Study – Developing ‘There is a Light: BRIGHTLIGHT’. Leeds: PROMS Annual UK Research Conference; 13 June 2019 (won the award for Innovation in PPI).

Taylor RM. Measuring Quality of Life in Young People with Cancer (Invited Presentation). Birmingham: TYAC Annual Conference; 11–12 September 2018.

Taylor RM, Lobel B, Thompson K, Croasdale M, Onashile A, McInally W, et al. There is a Light: BRIGHTLIGHT – Working with Young People to Create Impactful Dissemination (Oral). Birmingham: TYAC Annual Conference; 11–12 September 2018.

Taylor RM, Martins A, Lea S, Whelan JS, Fern LA. Participatory Research: Using Board Games to Guide Secondary Analysis of Big Data (Oral Presentation). Oxford: 2017 RCN International Research Society Conference; 6 April 2017.

Taylor R, Martins A, Lea S, Whelan J, Fern L. 003 PP: collaborating with young people to set the agenda for secondary analysis of the BRIGHTLIGHT dataset. BMJ Open 2017;7(Suppl. 2). https://doi.org/10.1136/bmjopen-2017-016492.21

Taylor RM on behalf of the BRIGHTLIGHT Team. What is BRIGHTLIGHT? (Oral Presentation). Liverpool: NCRI Annual Conference; 5–8 November 2017.

Taylor RM, Whelan JS. BRIGHTLIGHT: AYA programmes Adding Value (Invited Speakers). Atlanta, GA: 2nd Global AYA Cancer Congress; 5–7 December 2017.

Taylor RM, Martins A, Lea S, Whelan JS, Fern LA. Let’s Play Twister! Novel Methods to Facilitate Understanding of Complex Research Methods and Generate Secondary Research Hypotheses with Young People. Atlanta, GA: 2nd Global AYA Cancer Congress; 5–7 December 2017.

Taylor RM, Whelan J, Solanki A, Aslam N, Fern LA. ‘Okay to Ask’: Young People’s Views About Access to and Participation in Research (Oral Presentation). Sydney, NSW: Inaugural International Adolescent and Young Adult Oncology Congress; 3rd December 2015.

Taylor RM [symposium lead]. Teenagers, Young Adults and Access to Research: Perspectives from Policy, Participants and Researchers (Symposium Presentation). Glasgow: 2014 RCN International Research Society Conference; 4 April 2014.

Taylor RM [symposium lead]. BRIGHTLIGHT: A Lesson in Collaboration (Symposium Presentation). Glasgow: 2014 RCN International Research Society Conference; 3rd April 2014.

Taylor RM, Whelan J, Solanki A, Aslam N, Fern LA. Empowering Young People to Participate in Research: Young People’s Views About Research Access and Participation (Poster Presentation). Birmingham: NCIN Annual Conference; 9 June 2014 (received the patient choice award).

Taylor RM, Whelan J, Solanki A, Aslam N, Fern LA. ‘Okay to Ask’: Young People’s Views About Access to and Participation in Research (Poster Presentation). Liverpool: NCRI Cancer Conference; 4 November 2014.

Taylor RM, Fern L, O’Hara C, Solanki A, Whelan J on behalf of the BRIGHTLIGHT Team. A Multi-faceted Approach to Collaboration: The BRIGHTLIGHT Experience (Oral Presentation). London: Accepted Centre for Nursing and Midwifery Led Research: Research in Clinical Practice Conference 2013; 18 February 2013.

Taylor RM on behalf of the BRIGHTLIGHT Team. BRIGHTLIGHT: Do Specialist Services for TYA With Cancer Add Value? (Invited Presentation). Birmingham: TYA with Haematological Malignancy Conference; 28 March 2013.

Taylor RM, Whelan J, Solanki A, Aslam N, Fern L. The 7TH ‘D’: Don’t Decide For Me. (Poster Presentation). Leeds: TYAC Winter Education Day; 28 November 2013.

Whelan J, Pearce S, Jones L, Stirling C, Hough R, Riley V, et al. An introduction to BRIGHTLIGHT on end of life care for young adults: what do young adults with cancer and their families need and how can it best be delivered? BMJ Support Palliat Care 2013;3:133–4. https://doi.org/10.1136/bmjspcare-2013-000453b.25

Whelan J, Taylor R, Fern L. BRIGHTLIGHT: The 2012 TYA Cancer Cohort Study (Invited Presentation). Exeter, University of Exeter: South West TYA Conference; September 2012.

New website look

AS to lead: Show new web.

Ask to think if they like, remind them of their ideas from last time – are these represented on new website?

Load existing page to show new members of how the page currently appears. Ways to improve the general look of page.

Break for lunch 13.00–13.45

13.45–14.00 Lorna will explain secondary analysis and hypotheses generating.

Methods

Data presented for objective 2.2 have been chosen to reflect young people’s experience of the delivery of care (Table 2). Metrics have been proposed that are indicative of good TYA cancer care. 17–19,97–99 Metrics reflecting good TYA care at service and clinical record level are reported for objective 2.1 (e.g. considered for a clinical trial, discussed in a MDT), but a number of these service-level metrics were also reported through patient self-report, including fertility, access to members of the MDT and access to clinical trials. Questions were also included that related to the NICE guidance (choice of place of care) and reflecting age-appropriate care (involvement in treatment decisions, communication and co-ordination of care). 39,64 Some questions were asked at every wave of data collection and some were only asked at wave 1.

Results

Fewer young people who received ALL-TYA-PTC care were given the choice about where they could receive treatment and care, and fewer in the SOME-TYA-PTC group made a decision. The choice made by most of those in the NONE-TYA-PTC group was for care in the closest hospital, whereas more young people in the SOME-TYA-PTC group chose a hospital further away (Table 3). Irrespective of where young people were treated, the majority reported that health-care professionals had prepared them for treatment-related side effects (NONE-TYA-PTC, n = 256, 93%; SOME-TYA-PTC, n = 300, 96%; ALL-TYA-PTC, n = 187, 97%). Similarly, few young people reported that they would have liked to be more involved in their treatment decisions (NONE-TYA-PTC, n = 15, 5%; SOME-TYA-PTC, n = 15, 5%; ALL-TYA-PTC, n = 5, 3%). Young people in the NONE-TYA-PTC group had less access to members of the MDT representing specialist TYA services and supportive care than young people in the other two groups.

There were no differences in the number of young people who had spoken to someone about banking sperm or freezing eggs or embryos according to level of care; however, this occurred more often in males than females. Most young men were given the opportunity to bank sperm and mostly this was successful. More young women in the ALL-TYA-PTC group than the other two groups were given the opportunity to freeze eggs/embryos and took this opportunity (Table 4). Most young people felt the way their treatment team had handled the experience of fertility treatment was excellent/good (NONE-TYA-PTC, n = 93, 85%; SOME-TYA-PTC, n = 143, 88%; ALL-TYA-PTC, n = 91, 83%). Other reasons for deciding not to bank sperm or freeze eggs/embryos included receiving treatment that did not affect fertility, not yet reaching puberty and already pregnant when diagnosed. At every wave of data collection, fewer young people were offered entry into a clinical trial and fewer agreed to take part in the NONE-TYA-PTC group than the SOME-TYA-PTC and ALL-TYA-PTC groups (Table 5). Young people reported better communication from their nurse specialist than their cancer specialist (oncologist or haematologist) but there were no differences according to level of care (Table 6). The was also no difference in young people’s perceptions of the co-ordination of care (NONE-TYA-PTC, n = 246, 89%; SOME-TYA-PTC, n = 279, 90%; ALL-TYA-PTC, n = 183, 95%).

Methods

The BRIGHTLIGHT cohort included 1114 young people aged 13–24 years, who were newly diagnosed with cancer (ICD-10 codes C00–C97) in an English hospital and recruited within 4 months of diagnosis. Participants were identified through the national cancer waiting times data set and via tumour-specific MDTs at 97 NHS hospitals in England. Data from the HES database were used to calculate the costs associated with hospital care among study participants, including data for APC and outpatient visits (including chemotherapy and radiotherapy) and A&E attendances. Data for every person in the BRIGHTLIGHT cohort were sent to NHS Digital for linkage to the HES database covering all use of services over the period April 2012–March 2016. Linked HES data were then returned to the researchers for analysis.

To calculate hospital costs, we included all hospital contacts by participants in the BRIGHTLIGHT cohort, irrespective of cause, including chemotherapy and radiotherapy. For APC and outpatient visits we applied the relevant Healthcare Resource Group to each of our four financial years of HES data (i.e. 2012/13–2015/16) to produce a Healthcare Resource Group code for every admitted patient spell and outpatient attendance in that year. We then applied National Reference Costs from each financial year to calculate the cost of each spell and attendance at current prices. All costs were inflated to 2015/16 constant prices using the NHS Pay and Prices Index. 100 For A&E visits, we applied a single unit cost to every visit (£147.80), computed as the weighted average across all A&E visits in the National Reference Costs database. Data were recorded for every BRIGHTLIGHT patient on date of diagnosis, which was linked with the HES data based on the BRIGHTLIGHT study identifier. All APC spells, outpatient visits and A&E visits outside the first 12 months of diagnosis were dropped. For APC, this resulted in a data set of 20,215 costed spells across 1023 patients [mean cost per spell £1681, standard deviation (SD) £2994, median £681; interquartile range £368–2027]. For outpatients, there were 35,730 attendances across 1078 patients (mean cost per attendance £187, SD £180, median £160, interquartile range £129–220). For A&E attendances, there were 1221 costed visits across 574 patients. Patients with no costs for a specific type of contact (e.g. admitted care, outpatient, A&E) were assigned a zero cost for that type of contact.

Our main exposure variable was the categories of TYA specialist care (TYA-PTC) described in Quantifying specialist care. 63 We calculated mean 12-month combined hospital costs by categories of TYA specialist care and tested for significant differences using regression analyses. To account for skewness of the cost data, we used a generalised linear model with gamma family and log-link, running adjusted and unadjusted models. We also considered using log-normal, Gaussian, inverse Gaussian and negative binomial distributions, but the gamma model gave the best fit in terms of residual plots and the Akaike information criterion. In adjusted models we controlled for the same covariates used to analyse the primary outcome (see Identifying the outcomes associated with specialist teenage and young adult cancer care). 47 There were missing data for some of these covariates, so we reran the regression models including variables with no missing data only (i.e. disease severity,63 age at diagnosis, city/area, cancer type). We predicted hospital costs by categories of specialist TYA care controlling for the covariates (predictive margins).

Results

A total of 1044 patients were linked to the HES database with data on the categories of TYA specialist care (93% of the BRIGHTLIGHT cohort). Mean (SD) and median (interquartile range) costs of APC for the whole sample for the first 12 months from diagnosis were £30,473 (£36,599) and £15,528 (£6731–42,453), respectively (Table 7). For outpatient visits and A&E visits, the figures were £5472 (£5802) and £3649 (£1985–£6717), and £149 (£229) and £0 (£0–148), respectively. Costs were highly skewed (Figure 13). The combined mean and median costs were £36,094 (£38,835) and £20,365 (£9861–51,455); APC costs accounted for 84% of the mean combined cost, outpatient visits accounted for 15% and A&E visits accounted for < 1%. Of the 1044 patients in the sample, 359 (34%) had NONE-TYA-PTC care, 415 (40%) had SOME-TYA-PTC care and 270 (26%) had ALL-TYA-PTC care. For all four types of contact, unadjusted mean and median costs were highest among patients who received SOME-TYA-PTC care.

FIGURE 13.

Distribution of costs by type of hospital contact. (a) APC; (b) outpatient visits; (c) A&E visits; and (d) combined.

The covariates used in the regression analyses are described in Table 8. When adjusting only for covariates with no missing data (i.e. age at diagnosis, cancer type, severity group, city/area), mean hospital costs were highest among the patients receiving SOME-TYA-PTC care (mean £43,000, 95% CI £39,831 to £46,169) (Table 9 and Figure 14). There were no significant differences in costs between the NONE-TYA-PTC and SOME-TYA-PTC groups. The findings were broadly similar for the model including all covariates, with the highest cost incurred by the SOME-TYA-PTC group and then the ALL-TYA-PTC group, and lowest costs in the NONE TYA-PTC group. These differences were statistically significant.

FIGURE 14.

Predictive margins: (a) APC; (b) outpatient visits; (c) A&E visits; and (d) combined.

Methods

Participants were asked to complete a CoCQ retrospectively at 6 months after diagnosis, and cost records were completed prospectively at 6 months and 9 months for the following 3 months. Respondents were asked to record information relating to their travel costs and other out-of-pocket expenses related to the cancer and its treatment. The return rate for the questionnaires was 7–24% (Table 10). In terms of travel costs, participants were asked to record the names of up to five hospitals that they had visited since they were diagnosed with cancer and their usual method of transport to the hospital. The responses to the questionnaires were merged with the patient’s postcode of residence. The postcode of each hospital was found from online searches. The distance between the postcode of residence at diagnosis and hospital was computed using www.theaa.com, and travel costs were estimated according to the stated main method of transport. A total of 317 patients reported data for 966 trips (Table 11).

Methods of transport were car (84%), train (4%), taxi (3%), tube (3%), bus (2%,) on foot (2%), NHS transport (1%) and cycling (< 1%) (see Table 11). The last three of these were assigned zero cost to the patient/family. For car journeys, a cost per mile was applied from www.theaa.com; for taxis a cost per mile was applied from Uber (Uber Technologies, Inc., San Francisco, CA, USA); and for train journeys a cost per mile was applied from www.trainline.com. For buses and tube journeys, costs per trip were taken from www.tfl.gov.uk. All trips were costed as return journeys. Mean costs per trip by mode of transport are in Table 11. All car journeys were assumed to incur parking charges. The cost record recorded mean parking costs per week, which were applied to every APC stay. Parking costs for outpatient visits were based on national hospital parking costs per hour from the House of Commons, assuming a 3-hour stay. 101 Of the 966 trips, 494 were to a hospital with a TYA-PTC (mean travel cost £36.13), and 472 trips were to a hospital without a TYA-PTC (£27.77) (Table 12). We assumed that these were the mean travel costs incurred by patients and families to visit an ALL-TYA-PTC and NONE-TYA-PTC, respectively. These two mean costs were then applied to every outpatient visit and APC stay in the HES data over the 1-year period, differentiating between trips to a TYA-PTC or not. We included zero travel costs for A&E visits, assuming travel would usually be by NHS ambulance. Travel costs for APC trips were assumed to be incurred every day of the stay. Total travel costs were summed across all visits and stays per patient. These costs were analysed using the same statistical methods as for NHS costs.

For the other out-of-pocket expenses, patients and families were asked to record directly in Great British pounds the costs incurred under several headings over each time period: telephone calls, food, childcare, respite care, pet sitting/kennel fees, delivery/errand charges, cleaning services, domestic help, adaptations to the home, clothing, wigs, medical equipment, private medical fees, counselling and physiotherapy. Owing to small numbers, we only report mean values by categories of care.

Results

Unadjusted travel costs for SOME-TYA-PTC care were higher (£4168 per year, 95% CI £3836 to £4500) than those for ALL-TYA-PTC care (£2945, 95% CI £2654 to £3236) and NONE-TYA-PTC care (£1876, 95% CI £1708 to £2027). In the fully adjusted analysis, the sample size was 733 and the SOME-TYA-PTC group continued to incur the highest cost, followed by the ALL-TYA-PTC and NONE-TYA-PTC groups (Table 13). For the adjusted for covariates with no missing data, the trend was the same.

The mean total costs of out-of-pocket expenses for patients receiving NONE-TYA-PTC care, SOME-TYA-PTC care and ALL-TYA-PTC care for the first 6 months from diagnosis were £284.77, £743.83 and £976.46, respectively (Table 14). Similarly, out-of-pocket expenses for patients receiving NONE-TYA-PTC care, SOME-TYA-PTC care and ALL-TYA-PTC care at 6–9 months were £58.52, £280.58 and £122.14, respectively, and at 9–12 months were £398.66, 98.17 and 179.40, respectively. Food was the highest-cost item. Note that the low response rates decreased over time.

Methods

EuroQol-5 Dimensions, three-level version (EQ-5D-3L), questionnaires were completed at 6 months (wave 1), 12 months (wave 2), 18 months (wave 3), 24 months (wave 4) and 36 months (wave 5) post diagnosis and converted into utility scores using a UK value set. 102 Questionnaires were not completed at diagnosis as patients were not identified until up to 4 months after diagnosis. The QALYs were calculated for each respondent as the area under the curve up to 36 months. The values at wave 1 were assumed to have persisted since diagnosis. Response rates at each wave were 75%, 66%, 54%, 48%, 46% (Table 15), and QALYs were available for 18% of respondents, who completed all five waves, with no missing data. Utility scores at each time point and QALYs were analysed using the same statistical methods as for NHS costs. Given the extent of missing data, no imputation was undertaken, and in adjusted analyses we controlled for variables with no missing data only.

Results

The unadjusted mean QALYs per patient were 2.45 (95% CI 2.28 to 2.62) for patients receiving NONE-TYA-PTC care, 1.94 (95% CI 1.74 to 2.14) for those receiving SOME-TYA-PTC care and 2.20 (95% CI 1.94 to 2.28) for those receiving ALL-TYA-PTC care (Table 16). Trends were similar in adjusted analyses (Table 17), with overlapping 95% CIs between each group.

Cost-effectiveness of specialist care

The results for each separate analysis by category of care are summarised in Table 18. Bearing in mind the limitations of the analyses in terms of missing data and small numbers of responses, especially for the out-of-pocket expenses and QALYs, the NONE-TYA-PTC group persistently had the lowest NHS and travel costs of the three groups, and the highest QALYs. Notwithstanding the limitations of the data, taken at face value this suggests that SOME-TYA-PTC and ALL-TYA-PTC care are dominated by NONE-TYA-PTC care (i.e. are more costly and less effective). Given these findings, further long-term economic modelling was not undertaken.