Qualitative Analysis: Serious Illness Care Surveys

Qualitative researchers and UX teams should prioritize domain validity and use RAND-aligned phrasing when coding transcripts and designing instruments.

Prioritize domain validity: map codes to the RAND list of core domains to ensure content validity when you code transcripts or open-text survey responses.

Use cognitive-test-driven phrasing in instruments and probes to reduce respondent confusion and cross-walk existing datasets.

Compare telehealth vs in-person data explicitly, because RAND’s instrument wording was adjusted to cover both modalities.

Program leads and quality analysts should translate themes into measurable indicators for routine monitoring.

Translate themes into measurable indicators (for example, percent of respondents reporting timely access) for routine monitoring.

Use mixed-methods: combine patient-reported survey scores with qualitative quotes to explain trends to stakeholders and accrediting bodies.

Policy and accreditation teams should use standardized patient-reported measures for benchmarking across home-based and community programs.

Standardized patient-reported measures allow benchmarking across home-based and community programs, RAND provides an evidence-based starting point.

Document your cognitive testing and sampling (n counts above) to support measure reliability claims.

Evidano automates transcription with custom dictionaries and PII redaction, useful when RAND-style interviews include medical terms or family names.

Evidano can import the RAND domain codebook or create one from scratch and apply hierarchical codes to maintain consistency across n=85+ interviews or open-text survey responses.

Import the RAND domain codebook or create one from scratch; Evidano applies hierarchical codes→subcodes and AI-assisted coding to maintain consistency across n=85+ interviews or open-text survey responses.

Evidano runs thematic analysis, frequency counts, and cross-segment comparisons (for example, caregiver vs patient responses; telehealth vs in-person) in minutes rather than weeks.

Run thematic analysis, frequency counts, and cross-segment comparisons (e.g., caregiver vs patient responses; telehealth vs in-person) in minutes, not weeks.

Evidano generates word clouds, co-occurrence networks, and hierarchical code maps to make the RAND domains actionable for stakeholders.

Generate word clouds, co-occurrence networks, and hierarchical code maps to make the RAND domains actionable for stakeholders.

Evidano encrypts data end-to-end and does not use client data to train third-party models, suitable for sensitive patient-reported data.

A reminder for health research: Evidano encrypts data end-to-end and does not use client data to train third-party models, suitable for sensitive patient-reported data.

RAND published a methodological report that produces draft patient surveys for assessing serious illness care in home-based and community settings.

The report produced draft survey instruments after an environmental scan, an expert panel, in-depth interviews, provider interviews, and cognitive testing.

RAND conducted 23 in-depth interviews, 9 provider interviews, and 53 cognitive interviews as part of the study.

The in-depth interviews involved 23 seriously ill individuals and/or family caregivers, there were 9 health care provider interviews, and 53 cognitive interviews with seriously ill individuals and/or family caregivers.

RAND prioritized domains including communication; access/responsiveness; shared decision-making and advance care planning; symptom management; continuity/coordination; emotional/spiritual support; referrals for social determinants; caregiver support; and medication management.

Core domains prioritized are communication; access/responsiveness; shared decision-making & advance care planning; symptom management; continuity/coordination; emotional/spiritual support; social determinants referrals; caregiver support; medication management.

Researchers can reproduce RAND’s layered qualitative workflow quickly by combining automated transcription, cognitive-test-driven wording, codebook alignment, and AI-assisted coding.

Use automated transcription and custom dictionaries, import or create the RAND-based domain codebook, run AI-assisted initial coding, perform thematic and cross-segment analyses, and produce visuals and representative quotes for stakeholders.