Caregivers’ confidence in and preferences for reporting rehabilitation service use following paediatric traumatic brain injury: a qualitative study for measure development Authors: Jessica M Jarvis, Lauren Kaseman, Lauren Terhorst, Roxanna Bendixen, Amery Treble-Barna Published: 2026 ABSTRACT Abstract AbstractObjectivesTo evaluate caregivers’ confidence in accurately reporting rehabilitation service use for paediatric traumatic brain injury (TBI) and to determine their preferences for measure scope, granularity and administration.DesignThis qualitative descriptive study conducted remote, semistructured individual interviews with caregivers to examine their confidence in and preferences for reporting rehabilitation service use to inform measure development. The interview guide was refined through pilot interviews (n=2) excluded from analysis. Subsequent interviews were transcribed verbatim and analysed using thematic analysis with hybrid deductive-inductive coding.SettingHealth Insurance Portability and Accountability Act (HIPAA)-compliant videoconference platform.ParticipantsParticipants were recruited using maximum variation sampling to promote diversity across factors associated with rehabilitation service use. 14 caregivers met the following inclusion criteria and consented to participate: primary caregiver of a child aged 0–18 years who sustained a TBI requiring hospitalisation within the past 3 years; caregiver communicates in English. Sampling continued until data saturation was reached.ResultsThree themes pertaining to the study objectives were identified: (1) Caregivers are confident in their ability to accurately report their child’s rehabilitation service use and satisfaction by setting with approximate dosage; (2) Caregivers’ confidence in their ability to accurately report service use varies by setting and (3) Caregivers endorsed completing a comprehensive measure of rehabilitation service use with embedded definitions and structured prompts and preferred electronic administration for remote completion.ConclusionsFindings support the acceptability and potential feasibility of a caregiver-report measure of rehabilitation service use across the care continuum following their child’s TBI. Next steps will include prototyping the Rehabilitation Use Measure and iteratively refining it through expert feedback and cognitive interviews with caregivers.Keywords: Caregivers, QUALITATIVE RESEARCH, Health Services, REHABILITATION MEDICINE, PAEDIATRICS, Brain Injuries STRENGTHS AND LIMITATIONS OF THIS STUDY The interview guide was developed and refined through a systematic review of the literature, expert consultation and pilot-testing. The interview guide was developed and refined through a systematic review of the literature, expert consultation and pilot-testing. Maximum variation sampling was employed with the goal of capturing a broad range of perspectives from caregivers across key sociodemographic factors associated with rehabilitation service use, such as educational status, income and race. Maximum variation sampling was employed with the goal of capturing a broad range of perspectives from caregivers across key sociodemographic factors associated with rehabilitation service use, such as educational status, income and race. Despite recruitment efforts to support diversity across several demographic characteristics, most participating caregivers lived in metropolitan areas and were not Hispanic, and non-English speaking caregivers were excluded, which may limit the transferability of findings. Despite recruitment efforts to support diversity across several demographic characteristics, most participating caregivers lived in metropolitan areas and were not Hispanic, and non-English speaking caregivers were excluded, which may limit the transferability of findings. Rigorous thematic analysis was conducted using an iterative hybrid inductive-deductive coding approach with regular peer debriefing meetings among coders and senior investigators to review data, discuss emerging patterns, challenge assumptions and refine themes to support reflexivity and trustworthiness of findings. Rigorous thematic analysis was conducted using an iterative hybrid inductive-deductive coding approach with regular peer debriefing meetings among coders and senior investigators to review data, discuss emerging patterns, challenge assumptions and refine themes to support reflexivity and trustworthiness of findings. Introduction Paediatric traumatic brain injury (TBI) is a significant public health challenge, affecting over 630 000 children in the USA annually and leading to physical, cognitive and mental health impairments that can persist into adulthood.1,5 Functional recovery after TBI is highly variable and not well understood.6 7 Recovery models explain about a third of the variance in outcomes, impeding the delivery, evaluation and optimisation of therapeutic intervention.6 8 Recovery variability may be explained by differences in rehabilitation service use postinjury, such as the variety of service types, dosage and quality (eg, provider skill, family-centredness).9,11 A validated measure of rehabilitation service use that captures services across the care continuum, regardless of payer or healthcare system, would enable researchers to evaluate its contribution to recovery outcomes, account for services received when testing interventions and monitor service use patterns over time. However, collecting service use data is complex and resource intensive. Traditional methods, such as insurance claims or electronic medical records, are hindered by the diversity of payers, fragmented settings, absence of centralised medical records in the USA and lack of routinely capturing data related to the patient and family experience of rehabilitation services, such as extent of family engagement and their satisfaction with services. Caregiver report may offer a feasible approach for comprehensively capturing multiple dimensions of rehabilitation service use across the continuum of care. Our systematic review of the literature examining caregiver-report of rehabilitation service use after acquired brain injury identified only four measures with psychometric evidence for this population. These assessed experiential or quality-related components, such as satisfaction and care-delivery, to the exclusion of other dimensions of rehabilitation service use.12 These findings support the need for a validated caregiver-report measure of rehabilitation service use that comprehensively captures service use dimensions such as setting, service type, dosage and caregiver satisfaction. To address this need, we aimed to develop the Rehabilitation Use Measure to serve as a comprehensive, caregiver-report measure that quantifies the use of rehabilitation services. Best practices in measure development recommend qualitative inquiry to gather end-user perspectives to refine scope, inform item generation and support acceptability.13 This ensures the measure aligns with the caregivers’ lived experiences, preferences and perceived capabilities.14 The present study’s objectives were to evaluate caregivers’ confidence in their ability to accurately report dimensions of their child’s rehabilitation service use following TBI and determine their preferences for measure scope, granularity and administration. Methods Study design and participants This qualitative descriptive study used semistructured individual interviews. We used maximum variation sampling, a form of purposive sampling that seeks to include individuals representing a wide range of perspectives relative to the research question,15 to promote diversity across sociodemographic factors associated with rehabilitation service use, such as insurance status, race, ethnicity and urbanicity.16 To support this, we contacted over 70 brain injury associations and related organisations across all 50 US states and Puerto Rico to request support advertising this research opportunity; 20 organisations agreed, of which 9 were based in the Northeast. Enrolment was also advertised through social media and by contacting prior study participants. Interested caregivers contacted the study coordinator, who screened via telephone for the following inclusion criteria: primary caregiver of a child, aged 0–18 years, who sustained a TBI requiring hospitalisation; caregiver lives with the child >50% of the year; aged ≥18 years; and speaks and reads English. Caregivers of children who sustained a TBI >3 years prior to the interview were excluded to balance sufficient recruitment while limiting retrospective recall demands. The research adheres to the Consolidated Criteria for Reporting Qualitative Research guidelines (checklist available in the online supplementary materials).17 Data collection Interview guide development We developed the interview guide iteratively, drawing from Andersen’s Behavioural Model of Health Service Use and our systematic review, and revised in consultation with rehabilitation scientist and clinician experts and cognitive interviews with two caregivers.12 18 19 Cognitive interview data were not included in analyses. Guided by this process, we initially conceptualised rehabilitation service use as four dimensions of service use: setting, service type, dosage (session length, frequency and duration) and satisfaction (quality, sufficiency and caregiver engagement). These dimensions capture aspects of service use, distinct from contextual or individual factors that may predispose or enable service use (eg, service availability or TBI severity). Rehabilitation service types included functional therapies (occupational therapy, physical therapy and speech and language pathology), cognitive services (eg, neuropsychology), behavioural health services (eg, psychology) and community and school reintegration services (eg, school-based modifications and accommodations). The final interview guide, available in the supplement, was composed of open-ended questions with three sections: (1) caregivers’ knowledge of rehabilitation services; (2) their child’s rehabilitation service use following TBI and (3) priorities and preferences for measure scope, granularity and administration. Interview process Interviews were conducted by two members of the research team, JMJ and LK. JMJ, a music therapist and health services researcher with qualitative expertise, trained and mentored LK, a medical student with experience as a clinical research coordinator for paediatric TBI. Caregivers were first asked to name and describe all rehabilitation service types they were aware of. Unidentified services from our interview guide were named, and caregivers either confirmed knowledge and described them or received descriptions from the interviewer (see interview guide). This allowed us to ascertain familiarity across services, regardless of their child’s service use, and to identify key terminology or cues in their descriptions. Next, caregivers were asked to describe the rehabilitation services their child received after TBI, guided sequentially by setting: paediatric intensive care unit (PICU), hospital, inpatient rehabilitation, home, school and community. For each, they were asked to report service types, frequency, duration and satisfaction. Structured prompts elicited examples and explored caregivers’ confidence in their ability to accurately recall their child’s rehabilitation service use. Finally, caregivers were asked to provide their perspectives and preferences for a rehabilitation use measure, including scope, design and administration. Caregivers were interviewed once using HIPAA compliant Zoom videoconference platform (V.5.17.0)20 for 60–90 min. Interviews were audio recorded and transcribed verbatim, first using Otter.ai automatic transcription and then corrected and de-identified by members of the study team.21 Transcripts were not returned to caregivers and member checking was limited to the interview only, in which the interviewer periodically summarised caregivers’ responses to confirm accurate understanding. Analysis Transcripts were dually and independently coded by JMJ and LK, who also conducted the interviews, using NVivo V.20. Memos were encouraged to support reflexivity and training, but their content primarily pertained to training-related questions and noting potential exemplars and thus were not included in the analyses. We conducted a thematic analysis to identify themes pertaining to measure scope, granularity and administration using an iterative, hybrid inductive-deductive coding approach.22 23 Both coders applied the same a priori codebook, with two codes deductively applied to segments to indicate: (1) the service use dimensions (setting, type, dosage and satisfaction) and (2) caregivers’ reported confidence in accurately reporting this dimension; or a singular code to indicate the segment pertained to measure perspectives or preferences. Inductive codes developed pertained to factors influencing their reported confidence. Every 4–5 interviews, coding was compared and discrepancies resolved via discussion, which occasionally resulted in refining code definitions and recoding data. The a priori codebook remained stable following clarifications after the first eight interviews. To support reflexivity and trustworthiness, data and preliminary findings were reviewed at three points during data collection with the authorship team comprised of researchers external to data collection and with expertise in paediatric rehabilitation, measurement, neuropsychology, occupational therapy and qualitative methodology. Meetings served to review data and interpretation, challenge assumptions, build and refine themes and monitor for data saturation. Consistent with methods described for qualitative descriptive research, data saturation was a priori operationalised as the point at which sufficient data were collected to address study objectives and additional data did not contribute to new insights.24 Following the 12th interviews and the second team review, the authorship team noted that interviews were becoming redundant and saturation was being approached; two additional interviews were completed and reviewed, after which recruitment was determined complete. To complement the qualitative findings, we quantified caregiver familiarity with rehabilitation service types by counting the total number of services caregivers independently named and additional types they acknowledged familiarity with when prompted with the service name or description.25 Patient and public involvement The interview guide was refined through cognitive interviews with two caregivers of children with TBI, who provided feedback on question clarity and relevance prior to study initiation. These caregivers were compensated for their time. Caregivers were not otherwise involved in the study design, recruitment, or analysis. Results will be disseminated through peer-reviewed publication. Findings 16 caregivers were screened for study eligibility; 1 participant declined and 1 was ineligible due to their child’s TBI occurring >3 years prior. The remaining 14 caregivers consented to participate and completed their interview. Data saturation was achieved across this sample, representing 6 US states across all four Census regions, though predominantly the Northeast (10/14, 71%) (table 1). Caregivers’ characteristics varied with respect to their educational attainment and relationship to the child. Families predominantly lived in metropolitan areas (11/14, 79%), had public insurance (10/14, 71%) and were not Hispanic or Latino (13/14, 93%). Children’s hospital stays were distributed across short (2–3 days (5/14, 36%)), moderate (4–7 days (4/14, 29%)) and longer (>8 days (5/14, 36%)) durations. There was diversity in children’s age and race. Table 1Child and caregiver characteristicsVariablen (%)Child characteristicsAge at traumatic brain injury, years 0–56 (43) 6–114 (29) 12–184 (29)Time since injury, years <13 (21) 1 to <25 (36) 2–36 (43)Length of hospital stay, days 2 to 35 (36) 4 to 74 (29) >85 (36)Race American Indian or Alaska Native1 (7) Black or African American6 (43) White6 (43) Other1 (7)Ethnicity Not Hispanic or Latino13 (93) Hispanic or Latino1 (7)Caregiver characteristicsGender Female9 (64) Male5 (36)Education, highest degree earned High school diploma2 (14) Trade/technical certification4 (29) Associate degree1 (7) Bachelor’s degree4 (29) Graduate degree3 (21)Relationship to child Mother8 (57) Father3 (21) Other3 (21)Health insurance coverage Public10 (71) Private4 (29)Metropolitan classification Metropolitan11 (79) Non-metropolitan3 (21)United States Census Region Northeast10 (71) Midwest1 (7) South2 (14) West1 (7) Child and caregiver characteristics Caregiver knowledge of rehabilitation services Prior to discussing their child’s rehabilitation service use, caregivers were asked to name all rehabilitation service types they were familiar with. Most commonly, caregivers independently named physical therapy and speech and language therapy. Following prompting with unnamed service types and descriptions as needed, all caregivers confirmed knowledge of physical therapy, speech and language therapy, behavioural health and academic integration services. However, familiarity remained low with approximately half confirming familiarity with occupational therapy, neuropsychology and vocational services (figure 1). Unprompted, caregivers named a median (25th–75th percentile) of 2.0 (1.0–3.0) types of rehabilitation services; when prompted, caregivers confirmed familiarity with a median of 6.0 (5.0–6.8) of the 8 service types presented. Rehabilitation service types identified by caregivers. Thematic analysis yielded the following findings relating to a caregiver-report measure of rehabilitation service use: (1) Caregivers are confident in their ability to accurately report their child’s rehabilitation service use and satisfaction by setting with approximate dosage; (2) Caregivers’ confidence in their ability to accurately report service use varies by setting and (3) Caregivers endorsed completing a comprehensive measure of rehabilitation service use with embedded definitions and structured prompts, and preferred electronic administration for remote completion. Each theme is described below with representative quotes as exemplars of the caregivers’ feedback. Exemplar quotes are attributed to participants with the caregiver relationship, time since injury, and hospital length of stay (LOS), presented as ranges to protect participant confidentiality; additional descriptions of participant characteristics are available in online supplemental table S1. Caregivers are confident in their ability to accurately report their child’s rehabilitation service use and satisfaction by setting with approximate dosage. The majority of caregivers expressed confidence in their ability to accurately report whether their child received any rehabilitation services broadly (yes/no) within each setting and the approximate dosage. Caregivers also reported confidence in their ability to accurately recall their satisfaction with services for each setting, including their satisfaction with how they were engaged, the quality of care and whether services felt sufficient for their child’s needs. We did outpatient therapy with [place] for about 2 months He did physical therapy. And that was twice a week for an hour session. We did do Behavioral Health I want to say close to [9-months], he would go like once a month. – Mother, 1-<2 years post-TBI with 4–7 day LOS (#101) She just had her second neuropsych evaluation about a week ago. – Mother, 2-3 years post-TBI with >8 day LOS (#103) I feel like they did a really good job as far as keeping me informed, and keep me involved, and making sure that he good. I feel like they might have shot him too much at once. Especially as far as the questions — it’s like, you know, he just got out of the PICU and you’re asking him stuff like, ‘Who’s the football team in [city]’. Like, he’s — at the time, he was only like, seven — six or seven. So it’s like, how would he know that? But he did. So I mean, it shocked me [laughs]. – Mother, 1-<2 years post-TBI with 4–7 day LOS (#106) Across interviews, caregivers consistently offered estimates with qualifiers, such as ‘I think it was about’ and ‘around twice a week’, suggesting caregivers recall the general information on dosage while acknowledging some uncertainty in precise details. Still, most caregivers provided overall estimates for dosage as the number of sessions or typical duration by type within each setting, supporting potential feasibility of capturing dosage at a broad level. She was there [hospital] for a day and a half I want to say speech came in once to do swallowing. And then PT, OT came both days, I want to say. Yeah, because I remember them playing on the floor, and then kind of walking with her in the hallway a little bit. … I would say [each session] was a little longer than 15–20 min—around there.” – Mother, <1 year post-TBI with 2–3 day LOS (#108) [In the hospital] physical therapy and behavioral therapy. [Physical therapy] I wouldn’t say once a day, but I’d say once every, every other day. […] I am very confident. […] [behavioural health] Well, that was once every week. Basically, for about 40 to 45 min. – Father, 2–3 years post-TBI with>8 day LOS (#110) Caregivers’ confidence in their ability to accurately report dimensions of rehabilitation service use varied by setting Caregivers’ confidence reporting types and dosage was lowest for school-based services and highest for home and other community-based settings, such as outpatient clinics. Though some caregivers were confident in their ability to report their child’s school-based service use, the majority stated that their confidence in their ability to report detailed information on services received in the school setting was generally lower. When asked to describe the reasoning for setting-specific differences, caregivers attributed lower confidence reporting on service types and dosage for school-based services to insufficient opportunity to engage in services and limited communication with providers. In contrast, caregivers’ confidence in their ability to accurately report comprehensive details of service use across dimensions was highest for the home and other community settings, such as outpatient rehabilitation clinics. Caregivers shared that their active role in arranging and transporting their child bolstered their confidence in reporting accurately. Hmm. I really honestly don’t know much about what’s going on other than at the IEP meeting—she fills us in what she does. I just know how many hours he gets a month. I don’t know what they do. I don’t feel real involved. – Mother, 2–3 years post-TBI with 4–7 day LOS (#114) I remember it a lot because I […]took him to every appointment. – Father, 2–3 years post-TBI with>8 day LOS (#107) Caregivers’ confidence in their ability to accurately report their child’s rehabilitation service use in the hospital setting was the most variable among those interviewed. While many caregivers were confident reporting whether their child received rehabilitation services during their hospital admission (yes or no, broadly), their confidence in their ability to accurately report the specific type of service received was lowest in the hospital compared with other settings. […] They had like this book with, like, different pictures and stuff. So, I guess maybe it was something that was maybe her sight? I don’t know.” – Mother, 1-<2 years post-TBI with 2–3 day LOS (#104) I remember [in the hospital] at least one or two different people coming in from the rehab services, and then the various testing they would do with them. Physically and mentally. – Father, 2–3 years post-TBI with>8 day LOS (#107) Additionally, while the majority of caregivers readily reported whether their child was admitted to the PICU, confidence regarding service use within that unit varied. Some caregivers were confident their child did not start services in the PICU, often attributing this to their child’s clinical status at the time. I am sure it didn’t happen in the ICU. Because […] it was a period where he couldn’t actually do much […] it was when he started getting better, when he was, when he was coming out, when he was seeing the sun again, that’s when the whole physical therapy came on. — Father, 2–3 years post-TBI with>8 day LOS (#110) Others were less certain whether services began in the PICU, but remained confident reporting that services occurred during the hospitalisation broadly. I can’t remember if we transferred yet or not, though, but I know that she had three visits. – Father, <1 year post-TBI with 2–3 day LOS (#105) Well, again, I feel very confident that I know he received them. The specifics though, I just don’t remember. So I mean, it was probably pretty close to the PICU unit – Father, 2–3 years post-TBI with>8 day LOS (#107) Both caregivers with high and caregivers with low confidence reporting on service use in the PICU attributed their confidence level to the intensity of their experience with their child’s hospitalisation. Notably, this pattern was observed regardless of time since injury and length of hospital stay, suggesting that the emotional intensity of hospitalisation rather than recall period was the primary driver of confidence variability expressed in this setting. It’s possible that I could have lost a little detail. But I feel like, no, your child being hurt or being sick can be kind of traumatic, and that can really burn into your brain - that helps you remember, I guess. – Mother, 2–3 years post-TBI with>8 day LOS (#103) It was just the sheer number of people that were coming in, you know, doctors and nurses and […]we had social workers coming in, and there was just so many people in and out, […], there was a lot happening, which I think is why it becomes a little gray. –Mother, 2–3 years post-TBI with 2–3 day LOS (#102) Just the amount of people coming in and out and just still my mind being blown about what was going on and just lack of sleep.– Mother, <1 year post-TBI with 2–3 day LOS (#108) Caregivers endorsed completing a comprehensive measure of rehabilitation service use with embedded definitions and structured prompts and preferred electronic administration for remote completion Caregivers reported they would be willing to complete a measure of their child’s rehabilitation service use, citing altruistic motives when prompted, such as wanting to help future families and contribute to TBI research, with some indicating personal benefit from the process of reflecting. I feel like it’s beneficial. And you know, a reward of just being able to—we’re lucky with our outcome for our child, so I’d love to be able to, you know, help studies and see how that could help other people in the future. – Mother, 1-<2 years post-TBI with 4–7 day LOS (#101) In a good way bringing everything back, as far as you know, remembering like, oh yeah, we did do that or, you know, just things that we might not necessarily remember that occurred. — Mother, <1 year post-TBI with 2–3 day LOS (#108) Caregivers expressed a preference for the measure to comprehensively capture dimensions of rehabilitation service use. When asked which dimensions would be most important to include, many described all dimensions as contributing to a full understanding, with some explicitly resisting prioritisation and stating, ‘Everything will definitely be important’. When prompted to consider how the measure might be shortened, some suggested reducing the level of detail within domains, while others ranked dimensions, most commonly naming service types, dose and satisfaction as priorities. Notably, no caregiver recommended removing a dimension. Things like, ‘Were they in the PICU when this happened or were they on the regular floor?’ and ‘How long was each?’—those are ones that, for me, don’t jump out as much in terms of importance, I guess, for me, as do the ‘What types was she seen by and are you satisfied, is there anything that could have been done differently, how was the engagement?’’ – Mother, 2–3 years post-TBI with 2–3 day LOS (#102) […] what were the types, how long were those sessions, the amount of them, how engaged were you with them, and how satisfied you are with them – Mother, 2–3 years post-TBI with 2–3 day LOS (#109) I would just have to take the time to just do everything […], I feel like you need all of those pieces to really understand everything. – Mother, 1-<2 years post-TBI with 2–3 day LOS (#104) Caregivers’ preference for comprehensive capture informed their administration preferences. Most caregivers stated they preferred electronic format for remote completion, ‘for convenience’s sake’, and suggested ideal administration would take approximately 30 min. Their preferences for recall period in which they could maintain confidence in reporting accuracy varied widely, ranging from 2 months to 5 years. Caregivers also noted that the interview process improved their confidence in their ability to accurately report their child’s rehabilitation service use, particularly the opportunity to review names and descriptions of service types. The value of providing descriptions was also evident in patterns observed across interviews. Caregivers who initially reported limited familiarity with a service later recalled receiving it after its description had been provided. For example, when asked about neuropsychology during the initial service knowledge portion of the interview, one caregiver responded, “I think I’ve just heard of it” yet later when reporting their child’s service use independently stated “He received physical therapy, occupational therapy, speech, behavioral health. And he did have the evaluation for neuropsychology” (#101, Mother, 1-<2 years post-TBI). Caregivers noted that the interview process improved their confidence in accurately reporting, with one caregiver sharing “talking through it made me realize, yes, they actually did come into the room” (#109, Mother, 2–3 years post-TBI). Caregivers emphasised that reviewing service names and descriptions was helpful and that they should be embedded in the measure. Definitely the part where we familiarized on all of the kinds. That was very crucial I think doing it without that intro and without the familiarity of all of rehab services —it would be kind of challenging. — Father, <1 year post-TBI with 2–3 day LOS (#105) Discussion Unexplained variability in recovery after paediatric TBI is a critical barrier to clinical care and research progress. Rehabilitation service use may explain some of the variability; however, without a feasible, validated approach to measure service use across the care continuum, ability to evaluate these associations is limited.12 In this qualitative study to inform measure development, caregivers were confident in their ability to accurately report their child’s rehabilitation service use by setting, including approximate dosage and satisfaction. Caregivers endorsed willingness to complete a measure of their child’s rehabilitation service use and expressed a preference for comprehensively capturing dimensions of service use, with no caregiver recommending the removal of any dimension. Caregivers also preferred electronic, remote administration and noted that features of the interview process improved their confidence in accurately reporting and should be considered for measurement design. Here we discuss these findings within the context of future measure scope, design and administration. Rehabilitation service use is a multidimensional construct, and determining the appropriate scope for a caregiver-report measure is a critical early step in its development. Our recent systematic review revealed that existing, mostly unvalidated measures of rehabilitation service use after acquired brain injury predominantly focus on identifying a limited number of rehabilitation service types used, with few capturing aspects like dosage or setting of services.12 In contrast, findings from this present study demonstrate that caregivers are both confident in and prefer reporting service use dimensions comprehensively. This discrepancy indicates that the depth of data obtainable through caregiver reports may have been previously underestimated and underutilised. Our interview guide was designed to first establish caregivers’ familiarity with service types before reporting on their child’s service use across dimensions of service type, dosage and satisfaction by setting. Our findings supported and refined this framework, as caregivers confirmed these dimensions are important to capture and identified the setting of use as an important dimension that helps to organise their reporting. Furthermore, caregivers did not recommend removing dimensions, further supporting this comprehensive approach to measure scope. While caregivers identified all service use dimensions as important, they suggested these dimensions be captured broadly rather than at a granular level. For example, reporting settings of use (eg, hospital) but not setting specific units (eg, the PICU); reporting dosage as number of sessions but not session duration (eg, 20 min). This parallels the lower levels of confidence reported in their ability to accurately recall details such as the hospital unit-specific use and aligns with research indicating that recall is influenced by the meaningfulness of information.26 As a design specification, these findings suggest a rehabilitation service use measure should prioritise the breadth of service use information while calibrating the level of detail to what caregivers find meaningful and what they perceive they can accurately report, with the scope and granularity refined through future validity testing. Two findings from the interview process translate into concrete design specifications for the measure. Caregivers ultimately indicated familiarity with more service types when provided with descriptions versus an open-ended question or the names alone, and reported that reviewing service types increased their confidence in their reporting accuracy and should be incorporated into the measure. Caregivers also reported that the interview process, which used a forward chronologic recall with prompts and reflection, increased their confidence in reporting their child’s service use. It is important to emphasise that the confidence level reported by caregivers during the service use portion of the interview likely reflects a post-review rather than baseline confidence, as both the definitions and the reporting structure functioned as recall scaffolds. Therefore, these aspects should be intentionally replicated in the measure through embedded definitions, setting-specific service lists and sequenced item structure to support caregiver report. These design recommendations are consistent with findings from other resource-use measure development work, where cognitive interviews have similarly identified embedded definitions and examples as features that support comprehension of service categories.27 Caregivers primarily prefer remote, electronic measure administration for convenience and flexibility. This format would further support the integration of structured prompts and cues that would be impractical in a paper-based format. Electronic administration would support branching logic based on caregivers’ setting responses, limiting items to only those relevant to their child’s care trajectory and mitigating response burden. This aligns with reported benefits of integrating technology in rehabilitation and research, including greater accessibility, reduced burden and improved data accuracy.28,30 Although some caregivers felt confident reporting over a recall period of up to 5 years, evidence suggests shorter recall periods improve accuracy. In a study comparing self-report of rehabilitation service use against Medicare claims for a sample of 4228 older adults, accuracy declined from 90% with 3-month recall vs 60% with 9–12 months.31 Thus, a rehabilitation use measure may be best suited for prospective research designs with defined follow-up intervals to support shorter, bounded recall periods while enabling longitudinal data collection. Prioritising shorter recall periods for initial measurement testing may enhance data integrity, with future refinements optimising feasibility and reporting accuracy once validity is established. Limitations Despite broad national recruitment efforts, diversity across all targeted characteristics was not achieved. Our participants were commonly from metropolitan and Northeastern regions and were not Hispanic or Latino, partly reflecting the location of organisations that agreed to distribute recruitment materials and the exclusion of non-English-speaking caregivers. These factors may limit the transferability of our findings. However, we determined that continued recruitment was not warranted as data collected were sufficient to inform this preliminary stage of measure development. The 3-year inclusion window also created potential for recall bias and telescoping; however, confidence variability in our data appeared more closely related to individual responses to the stress of hospitalisation and caregiver involvement in service delivery than to time since injury. Additionally, member checking was not conducted, as we prioritised depth of initial accounts over participant verification of transcripts or findings. However, cognitive interviews in the next stage of measure development will provide a structured opportunity to seek caregiver perspectives on measure content derived from these findings. As the two researchers who conducted the interviews also coded the data, confirmation bias is a potential limitation; this was addressed through independent coding and regular peer review with the authorship team. Conclusions Caregivers expressed confidence in their ability to accurately report dimensions of their child’s rehabilitation service use after TBI and shared key insights on their perspectives and preferences for the measurement scope and administration. Findings support the acceptability and potential feasibility of a caregiver-report measure if designed to facilitate recall through embedded service definitions, structured scaffolding by setting and appropriate capture of service use dimensions without excessive detail. Future measure development steps currently underway include developing a prototype with rehabilitation scientist and clinician expert feedback, refining through cognitive interviews with caregivers and conducting reliability testing against administrative data. Supplementary material The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Footnotes Funding: Research reported in this publication was supported by the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) of the National Institutes of Health (Award numbers: R03HD101746 (PIs: AT-B and LT), K01HD097030 (PI: AT-B)). Prepublication history and additional supplemental material for this paper are available online. To view these files, please visit the journal online (https://doi.org/10.1136/bmjopen-2025-108351). Provenance and peer review: Not commissioned; externally peer reviewed. Patient consent for publication: Consent obtained directly from patient(s). Ethics approval: This study was approved by the University of Pittsburgh Institutional Review Board (IRB# STUDY21050111). Participants provided verbal informed consent prior to their participation and received a US$30 gift card. Data availability free text: The data (interview transcripts) are not publicly available due to privacy and ethical restrictions. Patient and public involvement: Patients and/or the public were involved in the design, or conduct, or reporting, or dissemination plans of this research. Refer to the Methods section for further details. Presented at: This work was previously presented as a poster at the American Congress of Rehabilitation Medicine, November 2023, Atlanta, GA. Data availability statement No data are available. References Supplementary Materials Data Availability Statement No data are available.