“It's very important that you measure throughout that journey…”: patient perspectives towards quality-of-life data collection following haematopoietic cell transplant

Abstract

Background

There is increasing interest in using Patient Reported Outcome Measures (PROMs) to provide evidence of how a haematopoietic stem cell transplant (HCT) affects blood cancer patients’ long-term quality-of-life (QoL). The purpose of this qualitative study was to explore patients’ opinions on what QoL data should be collected post-transplant, when this data should be collected, the use of this data beyond research, and how this data should be captured.

Methods

Twenty-one HCT patients with median age of 45 years (range: 26–71 years) took part in a semi-structured interview. Two commonly used PROMs, the Functional Assessment Cancer Therapy – Bone Marrow Transplant (FACT-BMT) and Patient Reported Outcome Measurement Information System (PROMIS-29) were used as discussion prompts. All interviews were audio recorded, transcribed verbatim and analysed in NVivo using thematic analysis. A Patient Advisory Group (PAG) (n = 6 patients) co-designed the study and were involved in reviewing the coding framework and findings generated.

Results

Patients expressed a strong preference for QoL data to be collected which is transplant specific and routinely captured over time. They felt QoL measurement could enable identification of post-transplant concerns, facilitate communication with health professionals, and facilitate access to personalised support. Many patients described being oblivious to the potential long-term implications of HCT and felt it would be reassuring to know the ‘typical’ trajectory of HCT recovery from patient outcome data collected > 100 days post-transplant. Patients were positive about electronic data capture but did acknowledge that depending on age, digital literacy, and access to electronic devices, a one-size-fits-all approach to QoL data collection would not suit all patients. Additional barriers to QoL measurement included the poor relevance and utility of the individual questionnaire items and concerns about whether PROMs were sensitive enough to capture day to day variation in wellbeing post-transplant.

Conclusions

Findings indicate patients are supportive of QoL data capture specific to transplant and feel such data could be used to support individual self-monitoring and post-transplant recovery. Patients feel that data should be collected routinely on a long-term basis via electronic methods.

Background

Haematopoietic cell transplantation (HCT) is an effective therapy for the management of blood cancers, non-malignant diseases, and bone marrow failure syndromes. Over 4,000 HCTs are performed in the UK each year [1]. There has been a reduction in morbidity and mortality post-transplant because of advancements made in the use of haploidentical donors, conditioning regimens, and effective post-transplant care. The HCT process is intensive, often leading to long-term impairments such as graft-versus-host disease (GvHD), allograft rejection, and infections [2]. In recognition of the burden of treatment on transplant recipients’ wellbeing, patient reported outcomes (PROs) are increasingly being used alongside traditional clinical outcome metrics [3,4,5].

Patient-reported outcome measures (PROMs) are defined as any outcomes related to the patient’s health or treatment that is evaluated directly by the patient, without any interpretation by a doctor or anyone else [6], and are used to collect PRO data, including quality-of-life (QoL) data [7]. The use of PROMs allows patient perspectives about the physical and psychosocial impact of disease and treatment to be measured. PROMs are usually developed, standardised, and validated with input from patients and clinicians. Initial questionnaire domains (i.e., general topics or concepts to be measured) are generated through literature reviews and discussion with patients. The draft PROM is then standardised through a series of quantitative studies which aim to psychometrically validate the questionnaire items and scales [8]. These validation studies ensure the final PROM accurately and consistently measures the target construct and captures information that is reflective of patients lived experience [9].

PROMs are defined as either condition specific or generic. Condition specific PROMs gather data which is specific to a particular patient group or condition. An example includes the Functional Assessment Cancer Therapy – Bone Marrow Transplant (FACT-BMT) which assesses the QoL of transplant recipients using the FACT-General (FACT-G) and a BMT-specific subscale [10]. Conversely, generic PROMs assess general health concepts which are relevant to a wide range of population groups. The Patient Reported Outcome Measurement Information System, PROMIS-29 is an example of a generic PROM which assesses QoL across 7 domains including anxiety, depression, fatigue, pain, physical function, sleep disturbance, and social participation [11].

Whilst feasibility of PRO data collection during the HCT pathway has been demonstrated [12], no study to date has investigated HCT patients’ views about which measure they think best represents their experience of how and when such data should be collected. As a result, little is known about the patient perspectives towards QoL data collection [13] and there is a lack of consensus about which PROMs are best to use in HCT research and when such data should be collected [14]. The purpose of this qualitative study was to explore patients’ opinions towards PROMs and their views on what QoL data should be collected post-transplant, when this data should be collected, the use of this data beyond research, and how this data should be captured.

Methods

Study design

This qualitative study used one-to-one semi-structured interviews to explore and understand patient views towards the collection of QoL data following HCT. The study was sponsored by Anthony Nolan, a UK charity which manages and recruits stem cell donors and provides support to HCT recipients. The study received approval from the Anthony Nolan Research Review Board (RRB) and adhered to the consolidated criteria for reporting qualitative research (COREQ) [15].

Patient involvement and co-design

A Patient Advisory Group (PAG) (n = 6), made up of individuals who had lived experience of receiving HCT, were involved in the design of the study materials and the interpretation of the results. PAG members attended two online meetings and two in-person workshops. The purpose of the two online meetings was to explore PAG views on key questions to investigate about quality-of-life data collection post-transplant. Following the online meetings, the research team (GP & KD) developed the study materials incorporating the PAG suggestions. Prior to the study, an in-person workshop was held to involve the PAG in co-designing the patient facing materials (patient information sheet, study advert and recruitment process) and test the draft interview guide. Following completion of study recruitment and data collection a second workshop was held and the initial codes and key themes from the study analysis were presented to the PAG to discuss. PAG members were involved in checking the validity, reliability, and relevance of the emergent themes.

Recruitment

Any patient ≥ 18 years of age who had received a HCT regardless of time since transplant was eligible to participate in the study. Participants were recruited via convenience sampling through Anthony Nolan channels. An open invitation to participate was circulated via Anthony Nolan hosted patient and family forums, channels and newsletters through a study advert, containing a link to register interest. Once the participant had completed the expression of interest form a member of the research team shared the participant information sheet, consent form and scheduled an interview time via email.

Semi-structured interview guide and data collection

Interviews for the study followed the semi-structured interview guide (Supplementary File A) and were conducted virtually via Microsoft Teams by a member of the research team (GP, KD or CY). During the interview participants were asked to share their opinion about two validated PROMs commonly used in research involving HCT recipients. The PROMIS-29 and FACT-BMT were chosen to be included within this study based upon existing literature​. Participants were sent both PROMs prior to the interview. These were then used as discussion prompts and participants were asked if they felt these measures reflected their experience of QoL post-transplant. During the interviews, participants were able to view copies of the PROMs via MS teams chat and were prompted to reflect on whether specific items within the measures resonated with their experience. Participants were also asked questions around what data they feel should be collected, when and how often, why this data should be collected and by what means. Interviews were continued until reaching data saturation, with no new themes emerging in subsequent interviews [16]. This article presents data on patient preferences on the use of PROMs for QoL data collection. Due to wealth of information collected, data on patient experience of QoL post-transplant is presented in a separate manuscript.

Qualitative analysis

Interviews were audio recorded and transcribed verbatim. Transcripts of the interviews were uploaded to NVivo 14 (QSR International) qualitative software and analysed using Braun & Clarke 6 phase thematic analysis [17]. Initial codes were generated by GP, who also led the analysis of interview material. KD undertook second coding. The research team (GP, KD, and CY) discussed codes and themes that were identified in the data. Once emergent themes were decided, quotations were selected from the transcripts to reflect main themes and sub-themes.

Incentive vouchers

All interview participants were offered a £25 incentive voucher. In line with NIHR funding guidance, the individuals in the PAG received voucher payments (£25p/h) as renumeration for their involvement in the study.

Results

Participant characteristics

Table 1 outlines the demographic and transplant characteristics of study participants. Thirty patients expressed an interest in taking part in the study, 2 dropped out due to illness ahead of the scheduled interview, 5 were excluded because they had not received a stem cell transplant in the UK and therefore did not meet the eligibility criteria, and 2 interviews were excluded from analysis due to poor audio quality. The majority of participants were female (57%, n = 12), White British (67%, n = 14), and over 45 years of age (71%, n = 15). The average time since HCT was 6 years (SD: ± 7 years). Almost a quarter (24%, n = 5) of participants were in receipt of social benefits, 29% (n = 6) had caring responsibilities and 48% (n = 10) lived rurally or > 2-h distance from a HCT centre.

Table 1 Participant characteristics

Qualitative interviews

All interviews were conducted in September 2023 via MS Teams, each interview averaged 55 min (range 32–82 min).

Emergent themes

Figure 1 outlines emergent themes and sub-themes relating to patients’ opinions towards PROMs and their views on what QoL data should be collected post-transplant, when this data should be collected, the use of this data beyond research, and how this data should be captured. Overall, patients preferred transplant specific data to be captured routinely over time and through electronic means. Patients felt PRO data could be used for to support individual self-monitoring and post-transplant recovery.

Fig. 1

Diagram of key themes and sub-themes emerging from interview material, discussed, and co-designed with patients (PAG)

Further quotations representing themes and sub-themes are presented in Table 2.

Table 2 Representative quotations from emergent themes and sub-themes

What data?

When asked to reflect on the FACT-BMT and PROMIS-29 questionnaires, most participants preferred FACT-BMT as a means of collecting PRO data. Patients felt the items included within the FACT-BMT were more reflective, relatable and in line with their lived experience of receiving a HCT. When reflecting on PROMIS-29, participants described it as ‘quite clinical and structured quite clinically or data-ry…’ [P17] (female, 26–39, 17 years post-transplant).

“The FACT-BMT is more open, you have more statements there that you can really identify from.” [P15] (female, 26–39, 1-year post-transplant).

Participants were positive overall about the use of FACT-BMT to collect transplant specific data on QoL. However, several patients queried whether PROMs are sensitive enough to capture day-to-day variation and a subtle change over time in relation to patient QoL post-transplant. Patients would often reflect when presented with the FACT-BMT and PROMIS-29 that the binary nature of the response scales don’t necessarily reflect their experience. For example, one patient commented about the fatigue response scale:

“Just because I can walk 15 min, doesn’t mean I don’t have fatigue. Just because I can do that, doesn’t mean that I don’t come home and have a nap straight away, because my body just needs to shut down for an hour” [P13] (male, 26–39, 3 years post-transplant).

Participants described the lack of detail they could give through PROMs on areas such as their return-to-work post-transplant, social isolation, the impact of being immunocompromised and the loss of independence and ability to carry out tasks as they would have pre-transplant.

“And the BMT questionnaire says I am able to work. Well, I am able to work. I’m not able to do my career that I was focused on for ten years before that. But I am able to work. So, my answer to that would be, I’m able to work quite a bit. Because that’s a factually correct answer. It doesn’t reflect the fact that I’ve had to give up my career. Because the job, and the career that I had before, I can’t do anymore. But I am, by that question, able to work. I could do a normal, nine-to-five job. That’s not what I did before, but I’m able to do that” [P13] (male, 26–39, 3 years post-transplant).

Why?

When asked why they think PROMs data should be captured post-transplant patients often responded that it would be helpful to monitor changes in their health as they recover. Participants described recovery following HCT as a ‘rollercoaster’ with some good days and some bad days. Many participants felt PROMs can allow for patients to have a better understanding of their own self-monitoring and wellbeing, as well as reassurance that they are not alone in their transplant journey.

“Well, in general, I think after transplant is an up and down thing, it's a rollercoaster. Sometimes you're doing good, sometimes you're not doing good, and being able to measure in which parts you really need the help, in which parts you have it sorted, is very helpful, because you can focus on what you need to do, or where do you need to ask for help…” [P15] (female, 26–39, 1 year post-transplant).

Most participants said retrospectively that they felt unprepared for what to expect during HCT recovery. Participants felt PROMs should be collected and used to raise awareness about life post-transplant, allowing new patients to feel more prepared for the intensity of recovery and potential HCT side-effects. For example, several participants noted that knowing there are others who have experienced similar symptoms post-transplant would have helped them feel less anxious and isolated:

“I didn’t have any awareness of what it could be like after treatment. Where if you could fill them in and somebody looks at them and then they say to you ‘oh well, that’s normal. You’re going to feel like that” [P4] (female, 55–64, 3 years post-transplant).

“You could use it to really inform patients as to what they are about to embark on… it won’t paint a particularly pretty picture, but it will paint a fair one” [P22] (female, 40–54, 8 years post-transplant).

Participants felt that collecting QoL data through PROMs could help ensure the right support is available following HCT. By completing transplant-specific PROMs, healthcare professionals (HCPs) could be made aware of patient symptoms and can use the data collected to guide what information and support they offer.

“Even shaping those conversations you have at follow-ups and bits like that as well or even then giving patients ideal awareness and signposting them to services by saying, ‘Oh we often find that patients experience this around this time. Don’t be alarmed about this, this is how we can help you’” [P17] (female, 26–39, 17 years post-transplant).

“They do need to know these extra things, definitely. Because I think it is about the whole package of support, it’s not just about the clinical. You do need to know what else there is out there, as well” [P12] (male, 65–71, 6 years post-transplant).

When?

Participants felt QoL data should be captured routinely, over a long period of time, not just until 100 days post-HCT (100 days is considered the timepoint when the greatest risk for critical side effects is past and when stem cells have engrafted). When asked to think about when QoL data should be collected many participants highlighted that data collected during initial post-transplant recovery (< 100 days) would be helpful to use as a baseline starting point for reflection.

“Day-zero makes sense because it’s kind of like a blank canvas, starting from scratch” [P2] (male, 26–39, 1 year post-transplant).

“Yes, I think that’d be really good, yes, because I think that when you go in for your transplant your quality of life is what you’re used to, and then I think straight after your transplant is when you’re still feeling quite ill and you’ve probably got the worst side effects, then 100 days you can look back and think ‘yes’. I think they’re quite good time scales, it gives you enough time to recover and to be able to look back and think ‘yeah, I have improved, yeah’” [P3 B2] (female, 40–54, < year post-transplant).

However, participants felt QoL data collected > 100 days post-transplant would be also useful as they felt this was often an overlooked part of their experience. They felt that long-term, continuous measures would help reflect various timepoints and capture change over time.

“I think it would be a good idea, because obviously I’ve filled it in a couple of months after my transplant, so I’m feeling a lot better now than I was before, so I think it would be a really good idea to fill it in just before you go in for your transplant, and then quite soon after coming out of hospital, and then now, and even going forward perhaps in another six months’ time, and then you see how your answers have changed… because I think sometimes when just day to day you think ‘oh things aren’t getting better’, but I think if you had a look back at the questionnaire a month ago I’d think ‘well actually yeah, I’m feeling a lot better in this area’… so I think it would be quite a good thing to be reflective” [P3] (female, 40–54, < year post-transplant).

How?

Overall, participants felt positive about electronic data capture and were positive towards the idea of receiving questionnaires on a tablet device or their phone during clinic visits, or at home.

“All for it. Yes, definitely. Anything that can prevent printing and paper waste. And I think nearly everyone now… I guess, 98% of your potential recipients would be able to complete it online” [P2] (male, 26–39, 1-year post-transplant).

Participants acknowledged that electronic data capture would be more convenient and would allow more certainty about the secure storage of the data. Some participants expressed concerns about their ability to complete ePROMs, with older participants claiming they felt more comfortable with a traditional pen and paper approach. Others shared thoughts on how this issue may be overcome if ePROMs were more prominent in clinic visits, where they would have access to support and guidance from health care professionals when completing the surveys.

“So, I think that old fashioned give it to me on a piece of paper and then I'll go home and look through all the bits of paper, but that might just be my age, young people would probably prefer the digital version because they're much better at tracking what they've responded to and what they haven't” [P1] (female, 40–54, 4 years post-transplant).

“I think particularly when we’re looking at multiple myeloma, generally, the age is older, so therefore perhaps not as digitally savvy, …perhaps need some help completing surveys. And again, perhaps that's where some of the support that I mentioned when you were in hospital with the nurses or from a team could assist people when they were in hospital to help complete those surveys” [P10] (female, 55–64, < 1-year post-transplant).

Participants also highlighted that transplant related side-effects may hinder their ability to complete PROMs electronically. For example, one participant noted the impact of GvHD on their eyesight and how this would have affected his ability to complete electronic PROMs.

“I think anybody above that is probably bordering towards more paper, from a vision perspective, if I think back to sometimes, I just couldn't see things online, I didn't use a computer for a long period of time because my vision was too bad with GvHD, I had whacking headaches, I couldn't even open my eyes” [P24] (male, 40–54, 2 years post-transplant).

Discussion

This study demonstrates that patients are generally positive about the collection of QoL data through transplant-specific PROMs. However, participants questioned whether existing PROMs are sensitive to capture day-to-day variation and subtle change over time in QoL post-transplant. Perspectives on why and when this QoL data should be collected were driven by a desire to know what to expect during transplant recovery, to learn from the experience of others, and to measure the long-term impact of transplant beyond 100 days. By capturing data > 100 days post-transplant, both patients and HCPs could observe symptom change over time, allowing for reflection and awareness of what support may be needed. Participants were positive about electronic data capture however variations in age, digital literacy and stage of illness may have a significant impact on preferences. Options should be given to patients on how to complete PROs to ensure effective data collection.

Previous qualitative work on PROMs has highlighted that often, when completing questionnaires following blood cancer, patients wonder if certain questions are relevant to them and their illness, and some struggle to identify with questions that do not relate to their situation [18, 19]. This study, whilst focused on patient experience post-transplant, shows similar findings when participants were presented with the PROMIS-29 and FACT-BMT. This study revealed that, despite FACT-BMT being the preferred option, patients still felt individual survey items may not fully capture the true patient experience of receiving a transplant. Several participants queried whether PROMs are sensitive enough to capture day-to-day variation and subtle changes over time. One of the benefits of PROs is that they give patients a voice, however these findings suggest that existing PROMs do not reflect patient experience of transplant to the extent that patients would like.

Participants felt QoL data should be collected to highlight patient experience. There has been a growing importance of the need to explore patient experience and the supportive care interventions needed to improve QoL post-transplant [20, 21]. Knowing what to expect post-transplant is essential in the decision-making process for patients when faced with a cancer diagnosis and the possibility of transplant [22]. In this study, participants expressed a desire to learn from others to gain more awareness about life post-transplant to be prepared for the changes to their QoL.

There is evidence to support the idea that routinely collected PROs with timely feedback enhances patient experience and satisfaction with care and enables HCPs to better understand and act on patient needs [23, 24]. Participants reflected on the benefit of capturing change over time. With the use of long-term, routine data capture, patients can reflect on how they felt < 100 days post-transplant to capture starting points and compare that to how they feel > 100 days post-transplant. This can act as a means of reassurance and motivation, as patients can visualise how far they have come in their recovery, as well as gaining awareness of what symptoms are normal, allowing them to raise any concerns with their HCPs.

Participants were unanimous that a ‘one size fits all’ approach to data capture is not appropriate and that multiple methods of data collection should be available to suit individual preferences. The idea of multi-modal methods of PRO data capture has been suggested in previous studies [25]. Participants shared concerns over the impact of age, digital literacy, and stage of illness in completing ePROMs. These are potential barriers to the use of ePROMs alongside large initial financial investment of purchasing a device to complete ePROMs [26, 27]. In this study participants highlighted concerns about the potential impact of GvHD on a patient’s eyesight, affecting their ability to complete PROs digitally. Whilst electronic data capture may be convenient for some patients and HCPs, other considerations should be made of individual circumstances [28]. Thought towards patient preference for PRO data collection should improve patient engagement with individual measures being used.

Strengths and limitations

Our study presents novel data on patients’ opinions towards PROMs and their views on what QoL data should be collected post-transplant, when this data should be collected, the use of this data beyond research. This study was developed with patient input through design and patients were involved in data triangulation. A convenience sample was recruited through Anthony Nolan patients and families’ channels: meaning a number of participants who live a distance from their transplant centre, receive benefits, and have caring responsibilities were able to take part.

Limitations of this study include potential subjective bias of patient experience and varying levels of health literacy which can lead to different interpretations and responses to specific questions. The use of convenience sampling may restrict full representation of patient population as some participants may have had a connection with Anthony Nolan.

Conclusion

This study has contributed to a significant gap in current literature by exploring patient preferences towards the collection of QoL data following HCT. Results of this study indicate that QoL data should be collected routinely, post-100 days to capture the long-term impact of HCT. Future research should pilot the prospective collection of electronic PRO data and should explore the feasibility of sharing data in real time with patients and health professionals to inform service delivery.