The Care-Full Study: assessing the feasibility of a mixed-method longitudinal data collection approach for unpaid carers of people with multiple long-term conditions

Common data collection methods for unpaid carer research

Research on unpaid carers in the UK has employed a mix of quantitative and qualitative methods to capture different facets of the caregiving experience. On the quantitative side, large-scale surveys are frequently used to measure key domains such as carers’ health status, quality of life, burden of care, and service use. For example, longitudinal panel studies like Understanding Society include modules where respondents report caregiving activities and their own health-related quality of life.⁵ In social care research, carer-specific outcome tools like the Adult Social Care Outcomes Toolkit for Carers (ASCOT-Carer)⁶ or the Adult Carer Quality of Life (Ac-QoL) instrument⁷ have been used to assess domains such as quality of life, psychological well-being, social isolation, and satisfaction with services. Routine national surveys (for example, the NHS biennial Survey of Adult Carers in England⁸) typically ask about carers’ quality of life, health, stress levels, hours of care, and experiences in navigating health and social care systems. These quantitative approaches yield broad indicators of carer outcomes and needs across large samples.

On the qualitative side, in-depth methods like interviews and focus groups are widely employed to capture the lived experience of caring, enabling carers to discuss challenges and share insights.⁴ More recently, participatory and co-design methods have gained traction. Carers are involved as active partners in research design or data collection – for example, nominal group techniques and workshops where carers help set research priorities or develop survey content.⁴ In some emerging research, carers (including former carers) have even been trained as peer researchers to conduct interviews or “check-in” conversations with other carers, leveraging their shared experience to build trust.^9,10 Such peer or co-research approaches acknowledge that carers’ insider perspective can enrich data collection and make participants more comfortable opening up.

Limitations and gaps in existing research

Despite the breadth of methods used, important limitations remain in how data on unpaid carers are collected. One major gap is the temporal infrequency of data collection in many studies. Much evidence about carers relies on cross-sectional snapshots or surveys administered at widely spaced intervals. For instance, national surveys or longitudinal cohort studies often collect carer data only at a single time point or at yearly/biennial waves, which can overlook the day-to-day realities of caregiving.^5,8 Caregiving for an older person with multiple conditions is dynamic – health crises, fluctuations in the cared-for person’s needs, or sudden changes in support can occur with little warning. Research that only samples experiences annually or sporadically will miss these important short-term variations. Similarly, retrospective questions that ask caregivers to recall events or caregiving intensity over long periods (e.g. “in the last 5 years”) are prone to recall bias and imprecision.¹¹ Moreover, classic qualitative data collection like interviews or focus group are rarely repeated over an extended period. These issues highlight that infrequent assessment or retrospective data collection with long recall period can fail to capture the true dynamics of caregiving experiences, potentially underestimating certain stresses or needs.

Another limitation is that carers’ lived experiences are not always adequately represented in either research design or the resulting data. Quantitative carer surveys typically focus on broad outcomes such as quality of life and wellbeing, but often lack the granularity needed to capture the fluctuating, day-to-day realities of caregiving. In contrast, qualitative methods such as in-depth interviews or focus groups can generate rich insights but are frequently time-intensive, making them impractical for carers who provide ongoing care without respite.⁴ Moreover, carers may feel more comfortable sharing their experiences in formats that are flexible, responsive to their individual circumstances, and facilitated by individuals with shared caregiving experience, which is not always achievable within conventional research approaches.^9,12,13

The current study

The limitations identified in existing carer research highlight the need for methodological innovation in how carers’ experiences are captured over time. In particular, there is a need for data collection approaches that are sufficiently frequent to reflect the dynamic nature of caregiving, flexible enough to minimise burden, and inclusive of carers’ lived experiences beyond what can be captured through standardised survey instruments alone. Advances in digital data collection, alongside growing recognition of the value of peer-research and participatory approaches, create new opportunities to address these challenges.

As a part of the larger Care-Full study (award ID: NIHR158213), we designed a mixed-method longitudinal data collection approach that combines regular, low-burden quantitative surveys with ongoing qualitative engagement facilitated by peer researchers. Specifically, the approach integrates fortnightly self-report surveys alongside monthly check-ins conducted by trained peer researchers with lived experience of caregiving. This design aims to balance the strengths of quantitative and qualitative methods: providing repeated, structured measures over time while also creating space for carers to reflect on and contextualise their experiences in their own words.

The primary aim of this paper is to assess the feasibility of this mixed-method approach, focusing on recruitment, retention, survey completion, missing data, and participant engagement over time, as well as the acceptability and perceived value of peer-researcher involvement. By reporting on these feasibility outcomes, the study seeks to inform the design of future longitudinal research with unpaid carers and contribute methodological insights into how mixed quantitative and qualitative data collection can be implemented in ways that are both rigorous and responsive to carers’ lived realities.

Patient and public involvement

Patients and members of the public, particularly unpaid carers of older people with multiple long-term conditions, were involved throughout the development and design of this study and continue to be actively involved in its conduct, governance, and dissemination.

When and how were patients/public first involved in the research?

Patient and Public Involvement and Engagement (PPIE) informed this study from its inception. The research design was directly shaped by learning from PPIE representatives and participants involved in previous NIHR-funded work within the same research programme, including the HEE/NIHR-funded PALLUP study (award ID: ICA-SCL-2018-04-ST2-001). These early discussions highlighted gaps in current support for unpaid carers, the complexity of caring for older people with multiple conditions, and the need for research approaches that better reflect carers’ real-world experiences.

How were the research question(s) developed and informed by their priorities, experience, and preferences?

The overarching research aims and questions were informed by carers’ priorities around recognising unpaid caring as a core component of home-based care systems, understanding key events and stress points in the caring journey, and identifying opportunities for earlier and more equitable support. PPIE input emphasised the importance of focusing on outcomes that matter to carers themselves, alongside outcomes for the older people they support, and the need to minimise burden while enabling meaningful participation over time. These perspectives directly shaped the study’s focus on mixed-method longitudinal data collection approaches that can reflect complexity and dynamic nature in unpaid caregiving.

How were patients/public involved in the design and conduct of the study?

Patients and the public were involved in multiple aspects of study design and conduct. PPIE representatives contributed to the development of participant-facing materials, including participant information sheets and consent processes, ensuring accessibility and clarity. The study governance structure includes PPIE representation through a Project Steering Group, providing oversight of study progress, ethics, and data management.

A distinctive feature of the study is the involvement of unpaid carers as peer researchers. Current or recently former unpaid carers were invited to take on this role and receive training and support to contribute to data collection and participant engagement for the longitudinal workstream. Their involvement supported both the research process and a more equitable and reciprocal research relationship.

How were patients/public involved in the choice of outcome measures?

The selection of outcomes and data to be collected was informed by participatory systems mapping workshops involving unpaid carers, carer advocates, and other stakeholders. These workshops enabled carers to identify outcomes they considered most meaningful, such as wellbeing, burden, sustainability of caring, and interactions with health and social care services. PPIE representatives also contributed feedback on proposed survey content and data collection tools, which were iteratively refined to ensure relevance, acceptability, and minimal burden.

How were patients/public involved in recruitment to the study?

Patients and public contributors supported recruitment through existing PPIE networks, voluntary and community organisations, and previous research cohorts where consent to re-contact had been given. Recruitment strategies were designed with PPIE input to be flexible and sensitive to carers’ time constraints, using a combination of open, snowball, and targeted approaches facilitated by trusted gatekeeper organisations.

How were patients/public be involved in choosing methods and agreeing plans for dissemination?

Patients and public contributors were involved in shaping the participatory and longitudinal methods used in the study, including the format, timing, and delivery of workshops and interviews. Plans for dissemination have also been informed by PPIE priorities, with an emphasis on sharing findings in accessible formats with participants and relevant communities. Study findings will be fed back to participants who wish to receive them through summary reports, and dissemination activities will include events involving carers, stakeholders, and partner organisations. Co-production with PPIE representatives will continue to guide decisions about how results are communicated and how learning from the study is shared beyond academic outputs.

Overview of the approach

This study employed a mixed-method approach to collect repeated quantitative and qualitative data from unpaid carers. The approach combined fortnightly self-report surveys with regular qualitative check-ins conducted by trained peer researchers.

Quantitative data were primarily collected through a study-specific digital survey hosted on Qualtrics and accessible via smartphone or web browser. This digital format was selected to maximise accessibility and flexibility and to reduce the logistical burden associated with repeated data collection. Survey content and design were developed iteratively in collaboration with a data consultation group and PPIE contributors, with at least two rounds of testing and refinement conducted with unpaid carers prior to feasibility testing. This process aimed to ensure clarity, relevance, and acceptability of survey items while keeping completion time as short as possible.

In addition to routine fortnightly surveys accessed by a specific link for each participant, the study incorporated an event-initiated data collection component. Participants were asked to notify the research team if a significant “point of change” occurred, such as a deterioration in the physical or mental health of the carer or the person being cared for (e.g. acute illness, falls, or hospitalisation). Following researcher review of the reported event, participants were invited to complete a series of brief daily micro-surveys for five consecutive days, covering a simplified subset of the fortnightly survey measures. This component was designed to capture short-term changes in carers’ wellbeing and caregiving demands during periods of heightened disruption.

Following survey development, the full data collection protocol was tested for feasibility with 15 unpaid carers. Feasibility outcomes included recruitment and retention, survey completion rates, non-response, and missing data patterns.

Recruitment of participants

Participants were recruited across three study sites in England (West Sussex, South London, and Bradford) using a multi-pronged strategy designed to maximise reach and reduce burden for unpaid carers. Recruitment routes included open advertisement via regional and national voluntary and community sector organisations, snowball sampling through existing participants, and targeted recruitment from pre-identified cohorts from previous and ongoing studies where consent to re-contact had been provided. Local gatekeepers played a key role in site-specific recruitment, with formal gatekeeper approval obtained prior to recruitment through the following organisations: Carers Support (West Sussex), St Christopher’s Hospice (South London), and Health Action Local Engagement (Bradford).

Participants who expressed an interest in participating in the study via email to the study team were sent the participant information sheet and invited to participate in a screening call, where they were informed about the study, and their eligibility and suitability was assessed against the inclusion criteria. Following a successful screening these potential participants were sent details for joining a baseline assessment and the link to an online consent form for completion at the beginning of the baseline assessment call or prior to the call if they wished.

Participants were offered remuneration in recognition of their time and contribution, in line with NIHR guidance. Participants taking part in participatory mapping workshops and interviews received a £50 gift voucher. Members of the longitudinal cohort were remunerated over the 12-month study period with an initial £30 voucher followed by monthly £20 vouchers (total £270). Participants were also offered two coaching sessions regarding their unpaid caring responsibilities with a trained coach. The voucher and coaching offers were optional and reflected the extended and ongoing nature of participants’ involvement in this longitudinal study. This approach aimed to promote reciprocity, equity, and sustained engagement across all participant groups.

Baseline assessment

Baseline data were collected at the start of the study to capture carers’ broader personal, social, and caregiving contexts. Baseline assessments were conducted via an online meeting with a member of the research team, according to participant preference, to support engagement and continuity in the study.

The baseline assessment included qualitative measures and incorporated an eco-mapping exercise. During eco-mapping, participants were invited to draw and describe a visual map of key relationships with individuals, services, and organisations involved in their caregiving lives. These data were used to contextualise longitudinal survey responses, support interpretation of changes over time, and enable the development of individual case narratives across the study period.

The baseline assessment also included an online survey covering all domains assessed in the longitudinal surveys (see below for details), alongside the collection of key background characteristics. These included demographic information (age group, gender, ethnicity, highest educational attainment, marital status, and employment status), prior caring experience, the relationship to the person being cared for, and current living arrangements. In addition, participants reported their use of formal and informal carer support services over the previous 12 months. The list of services was adapted from the Personal Social Services Survey of Adult Carers in England¹⁴ to ensure coverage of support types relevant within the UK care system.

Design of longitudinal survey

The longitudinal survey was informed by established survey methodology and drew on validated measures widely used in caregiving and health research. Measures were selected to balance psychometric robustness with feasibility for repeated administration. Key measures included in the survey are described below.

Quality of Life

Carer quality of life was assessed using the Adult Carer Quality of Life (AC-QoL) questionnaire.⁷ The AC-QoL measures eight domains reflecting both positive and negative aspects of caregiving: support for caring, caring choice, caring stress, financial implications, personal growth, sense of value, ability to care, and carer satisfaction. Each domain is assessed using multiple four-point Likert-scale items referring to experiences over the previous two weeks. Prior research has demonstrated good internal consistency and construct validity,⁷ including satisfactory convergent and discriminant validity with measures of burden, resilience, satisfaction, and physical and mental health, supporting its use among family carers.

Health and Wellbeing

General health status was assessed using a single-item self-rated health question with five response categories (excellent, good, average, poor, terrible). Symptoms of depression were measured using the eight-item Patient Health Questionnaire (PHQ-8),¹⁵ which assesses core symptoms of major depression over the previous two weeks. Additional wellbeing measures included self-reported average sleep duration and sleep quality over the past two weeks, and engagement in physical activity during the past week.

Caregiving Activities and Work Impact

Caregiving activities were grouped into three domains:

For each domain, participants reported typical time spent on activities over the previous three days, as well as the perceived mental and physical demands of these tasks during the same period. Work participation was assessed by asking participants to report the number of hours worked during the past week, hours of work missed due to caregiving or other reasons, and perceived impact of caregiving on productivity. These items were adapted from the Work Productivity and Activity Impairment Questionnaire: Informal Caregiving (WPAI:CG)¹⁶ and elements of the NASA Task Load Index (NASA-TLX).¹⁷

Monthly peer researcher check-ins

Peer researchers, referred to as Care-Full Companions, who were current or former unpaid carers recruited through existing PPIE contacts from past research projects, were supported through training and supervision and paid remunerated at an hourly rate aligned with NIHR payment guidance for public involvement in research.

The role of these peer researchers was to conduct monthly check-ins with participants. Three peer researchers (two female, one male) were purposively recruited for prior experience in unpaid caring and interest in supporting research. Public and Patient Involvement and Engagement (PPIE) contacts from past research projects who were known to have the necessary skills and experience were contacted and given study information including a role descriptor to gauge interest. Informal discussions were held with the study team and perspective peer researchers could discuss the role with the study PPIE lead. Contracts were arranged for peer researchers’ paid involvement in the study as described in the ‘Recruitment of Participants’ section further above.

Peer researchers attended an in-person training day, facilitated by the PI and PPIE lead. Content of the day included an in-depth discussion of their role, working together as peer researchers and how they wished to meet and support each other throughout data collection. Active listening skills were shared and practiced. A topic guide and checklist to guide monthly check ins was co-produced during and after the initial full training day with peer researchers, to facilitate consistent data capture while allowing flexibility to respond to participants’ individual circumstances. A companion reflexive guide was also developed to aid reflections and journaling after each check-in.

Each peer researcher conducted monthly check-ins with five participants via telephone or Microsoft Teams. The post-doctoral research associate (PDRA) set up the meetings and attended the beginning of each meeting. This ensured that all could attend, hear and the conversation was recorded and saved directly to University of Surrey secure servers.

Monthly conversations served multiple purposes: supporting participant engagement, providing opportunities to expand on survey responses, identifying experiences not captured through structured measures, and assessing the acceptability of peer-facilitated qualitative data collection over time.

Peer researchers could contact the team via the PDRA at any point for urgent concerns and to obtain responses to participant questions at the monthly check-ins. Peer researchers met together quarterly over the duration of data collection (3 times at Months 3, 6, and 9) to reflect on their experiences as study companions and to provide feedback to the study team. These meetings supported integration of peer researchers experience and expertise into the wider study team activities. A final meeting post-data collection between the peer researchers and investigators (FS, CN, and RG) was an opportunity to collect overall feedback on peer researchers’ experiences during the project.

Ethics

This study was given favorable ethical opinion from the University of Surrey research ethics committee (Ref: FHMS 23-24 005 EGA).

Participant characteristics

Following the initial screening call with the PDRA, eligible participants who wanted to take part in the study were invited to provide their written informed consent and their availability for the study baseline visit. Fifteen unpaid carers were recruited, and baseline visits completed, during a 15-week period between February and May 2024 (eight participants completed the baseline visit in February, five in March-April, and two in May).

Table 1 summarises the characteristics of the 15 carers who participated in the feasibility study. The sample was predominantly female and largely comprised individuals from White ethnic backgrounds, with representation from several minority ethnic groups. Most participants were middle-aged, with a smaller proportion of younger and older carers included. Educational attainment was generally high, with around two-thirds reporting higher education qualifications. Over half of the sample were married or cohabiting and a similar proportion were in paid employment, although a notable minority were either unable to work or not currently seeking employment. Prior caring experience was mixed, indicating variation in familiarity with caring roles among participants. In most cases, the person being cared for lived with the carer, while others supported individuals living independently or in supported accommodation. Overall, the sample reflects a diverse range of sociodemographic and caring contexts, supporting the feasibility of the mixed-methods data collection approach across varied carer circumstances.

Retention, completion rate, non-responses and missingness pattern

Recruitment took place between February and May 2024. All participants completed the baseline survey at enrolment and subsequently took part in the planned bi-weekly longitudinal surveys. In line with the study protocol, all data collection concluded in mid-December 2024. No participants formally withdrew from the study. However, three participants (20% of all participants) discontinued participation during the follow-up period due to the death of the person they were caring for. Overall, retention was high across the study period, with attrition attributable solely to bereavement rather than disengagement with the study procedures.

Completion of the longitudinal surveys was high over the study period. Among the three participants who discontinued participation due to bereavement, completion ranged from one to four waves prior to discontinuation. The remaining twelve participants completed between nine and seventeen survey waves, reflecting variation in the timing of study entry. Overall, 189 longitudinal survey requests were issued, of which 176 were completed; 13 surveys were either not responded to or were initiated but not completed, yielding an overall completion rate of 93.12%. When restricted to participants who remained in the study until completion, 168 of 179 requested surveys were completed, corresponding to a completion rate of 93.85%. These findings indicate strong engagement with the longitudinal data collection procedures over an extended follow-up period.

Overall, missing data was minimal, with an overall missingness rate of 0.7% across the study variables. The highest levels of missingness were observed for the two questions assessing the physical and mental burden associated with assisting the daily living activities of the care recipient, each with a missingness rate of 6.8%. In contrast, items assessing carers’ quality of life, depressive symptoms, sleep, and self-rated health exhibited very low levels of missingness (≤0.5%).

We tested the feasibility of capturing acute “points of change” in carers’ circumstances over a four-month testing period. Overall, there were seven notifications received. Of these, two were deemed not to represent genuine points of change (e.g., reporting a grandchild’s holiday break), and two were submitted several days after the event had occurred. Three notifications were judged by the research team to constitute true points of change and triggered a five-day follow-up survey protocol. However, this approach proved challenging in practice. Participants frequently expressed uncertainty about what constituted a qualifying point of change, and the events of interest themselves often coincided with periods of heightened stress and disruption to daily routines. This contributed to delayed reporting and a lower completion rate for follow-up surveys (66.67%) compared with the routine fortnightly longitudinal surveys (93.12%). The research team additionally encountered operational challenges in administering follow-up surveys, including notifications occurring outside standard working hours (e.g., weekends). Given these feasibility and burden-related challenges, testing of the point-of-change survey component was suspended after four months.

Participation in Monthly peer-researcher check-ins

Monthly check-ins with the peer researchers commenced approximately 4 weeks after baseline and continued on a 4-weekly basis thereafter, where possible. One participant met with their peer researchers on a 6-weekly basis.

The maximum number of check-ins during the active study period was ten, although the total number available varied depending on when participants joined the study. Completion rates were high, i.e., the frequency of the monthly meetings was maintained with the following exceptions: 5 participants missed a single check-in for carer-related reasons, 3 participants missed 1 or 2 check-ins for personal reasons, and 1 participant had a two-month gap between their second and third check-ins following bereavement. The third check-in has been at their request by way of follow-up with their peer researchers.

Each check-in meeting was arranged for one hour, however, the duration of individual check-ins varied by carer, their caregiving responsibilities and life events more generally at any specific time. Some check-ins got shorter as the participant became more familiar with the survey and with little or no change in their circumstances to share. Whereas others took advantage of the full hour and used it to ‘offload’ with someone outside their usual support network.

A total of 11 participants took part in the optional, post-study “exit interview” to provide their feedback on their study participation; specifically, in relation to data collection and the role of the peer researchers. Only participants who were still caring and providing study data in December were invited to take part in the exit interview.

Feedback

Feedback on the research and data collection approaches collected through monthly check-ins and the exit interview and survey are summarised below:

Getting involved and commitment to the research

Participants described a strong motivation for getting involved in the study, most commonly linked to an interest in research and a desire to contribute to improving understanding and support for unpaid carers. Many participants expressed a general belief in the value of research, with some noting prior involvement in PPIE or research activities and a sense that carers’ voices are often underrepresented. For several, the study “came at the right time,” offering an opportunity to reflect on their caring role during a particularly demanding period. Continued participation was often framed as a matter of personal commitment once enrolled, captured by sentiments such as “I’ve started, so I’ll finish.” While financial incentives were appreciated, participants consistently emphasised that these were not the primary motivator for engagement; instead, the perceived relevance of the research and the opportunity to be heard were central to sustained involvement.

Accessibility

Feedback indicated that the study was generally perceived as accessible and manageable alongside caring responsibilities. Participants valued the flexibility of participation, particularly the ability to complete surveys at a time that suited them and to engage in check-ins remotely. The digital survey format was described as straightforward and easy to navigate, even during periods of heightened stress. Clear communication from the research team and reminders were seen as supportive rather than burdensome. Some participants noted that carers with different levels of digital confidence or from underrepresented groups might experience greater barriers, suggesting that additional modes of access or tailored support could further enhance inclusivity. Overall, however, the approach was viewed as appropriately sensitive to the constraints faced by unpaid carers.

Data collection approach

Participants reflected positively on the overall balance between structured quantitative surveys and more open-ended opportunities to share experiences. The fortnightly surveys were consistently described as quick to complete, manageable alongside caring responsibilities, and broadly relevant to their lived experiences. Several participants noted that the regular survey schedule prompted reflection on aspects of wellbeing and caregiving that might otherwise go unnoticed, particularly during periods where routines became normalised despite ongoing strain.

Importantly, participants valued the opportunity to contextualise and elaborate on survey responses during the monthly peer-researcher check-ins, especially when fixed-response items felt insufficient to capture the complexity or emotional texture of their experiences. The ability to explain why a response had changed, or why it had not, was seen as critical to making the data collection feel meaningful rather than mechanistic. During the first four months of the study, small iterative refinements were made to the survey content and wording in response to participant feedback. These changes, including clearer phrasing and the inclusion of additional free-text options, were welcomed and reinforced participants’ perceptions that the study was responsive to their input.

As mentioned above, participants and researchers both identified challenges with the event-triggered “point of change” survey component. Participants reported uncertainty about what constituted a qualifying event, and periods in which meaningful changes occurred were frequently described as times of heightened stress and reduced capacity to engage with additional study tasks. This contributed to delayed reporting of events and lower completion of follow-up micro-surveys compared with routine fortnightly surveys.

The monthly peer-researcher check-ins were consistently highlighted as a particularly valued component of the study. Participants emphasised the importance of continuity, with many expressing a strong preference for meeting with the same peer researcher each month in order to build rapport and trust. The shared lived experience of caring was seen as central to the success of the check-ins, enabling participants to speak openly without feeling the need to explain or justify their feelings. Participants described the check-ins as supportive, reflective, and distinct from formal services, with one noting that the companion “understood without me having to explain everything.” While the primary purpose was not therapeutic, participants often experienced the conversations as emotionally validating and helpful in making sense of their caring role. A small number reflected on the possibility that different personalities might not always align, but overall, the peer-researcher model was viewed as acceptable, appropriate, and integral to sustained engagement in the study.