Evidano home
Commentary on News

What HCV Access Research Shows About AI Qualitative Analysis

Evidano9 min read

Healthcare researchers need more than percentages to understand why patients delay or avoid treatment. The qualitative analysis of healthcare interviews explains how language, cost, transportation, trust, stigma, and fragmented care shape decisions that administrative data alone cannot show. A November 2026 study of Hepatitis C virus treatment access in Los Angeles County demonstrates the value of combining program data with patient interviews. This article refracts that study through AI qualitative analysis in healthcare, showing how research teams can organize transcripts, compare treated and untreated participants, and connect themes to measurable barriers. The goal is not to replace clinical judgment or human interpretation. The goal is to help researchers move from large volumes of qualitative data to a defensible, transparent understanding of patient experience. The study also provides a practical model for public health teams investigating treatment gaps among underserved populations.

Key Takeaways

According to the PLoS One study on Hepatitis C treatment access, AI qualitative analysis in healthcare can help researchers connect reported barriers with the lived experiences behind those barriers. The Los Angeles County study found that personal, provider-related, insurance, and medical-access barriers interacted with language, housing, incarceration, addiction, and trust.

  • Between October 2024 and January 2025, the study analyzed de-identified program data from 344 Los Angeles County case-residents and interviewed 10 participants.
  • The study reported that 54% of case-residents experienced barriers, while personal barriers represented 43%, provider-related barriers 31%, insurance barriers 13%, and inadequate medical access 13% among participants reporting specific barriers.
  • The qualitative sample included 7 untreated participants and 3 treated participants, enabling comparison of barriers before treatment with factors that supported treatment completion.
  • The study identified language barriers, unclear information, cost, transportation, long waits, homelessness, incarceration, and drug addiction as barriers, while motivation, family support, provider trust, insurance, and treatment knowledge facilitated access.
  • The authors recommended language support, patient navigation, care coordination, mobile clinics for people experiencing homelessness, continuity after incarceration, and fewer insurance restrictions.

How the Los Angeles County HCV Study Used Mixed Methods

According to the November 20, 2026 PLoS One study, the research team used a convergent parallel mixed-methods design to analyze quantitative program data and qualitative interviews during the same October 2024 to January 2025 period.

The quantitative component used descriptive frequencies and percentages from Project HCV Connect, a linkage-to-cure program supported by the Los Angeles County Department of Public Health and the University of Southern California. The researchers did not conduct inferential modeling because most specific barrier categories contained fewer than 15 observations.

The qualitative component used semi-structured interviews conducted in English and Spanish. According to the study, interviews lasted 30 to 60 minutes, were transcribed verbatim, and were analyzed through iterative inductive coding. Team members independently coded transcripts, compared codes, and resolved discrepancies through discussion.

The study's design shows why AI qualitative analysis should support, rather than obscure, research reasoning. An AI system can accelerate transcript organization, code retrieval, thematic comparison, and cross-segment analysis, but researchers still need to define the research question, review interpretations, protect participant privacy, and report sampling limitations.

One untreated Spanish-speaking participant described the communication barrier directly: "Since I don’t speak 100% English... I didn’t understand anything." The quotation illustrates why frequency counts alone cannot explain how language affects treatment engagement.

HCV Treatment Access Findings at a Glance

Date or study measureMetricValueImplication for qualitative research
October 2024 to January 2025Case-residents contacted344A large program dataset can identify the prevalence and distribution of reported barriers.
October 2024 to January 2025Case-residents already treated168, or 49%Treatment status creates a useful segment for comparing completed and incomplete care journeys.
October 2024 to January 2025Case-residents not treated176Untreated participants can help researchers examine delays, uncertainty, and access failures.
October 2024 to January 2025Case-residents reporting barriers54%More than half of the program population reported an obstacle to treatment access.
October 2024 to January 2025Interviews completed10, including 7 untreated and 3 treated participantsA small qualitative sample adds context that administrative categories cannot provide.
October 2024 to January 2025Specific barrier distribution among those reporting specific barriersPersonal 43%, provider-related 31%, insurance 13%, inadequate medical access 13%Thematic analysis can explain what each category means in patients' actual care experiences.
2019 to 2023People diagnosed with HCV in Los Angeles County27, 678The scale of diagnosis creates a substantial need for scalable linkage-to-care research.
Reported in the November 20, 2026 studyPeople reporting knowledge of HCV treatment30%Patient education is a major research and intervention opportunity.
Reported in the November 20, 2026 studyPeople completing HCV treatment32%Low completion rates make it important to study both barriers and facilitators.

What AI Qualitative Analysis Adds for Healthcare Researchers

How can AI help researchers find barriers across healthcare interviews?

AI qualitative analysis can identify recurring themes across interview transcripts while preserving the context of each participant's account. Researchers can use thematic and frequency analyses to compare references to language, cost, transportation, stigma, provider trust, and treatment knowledge.

The November 20, 2026 PLoS One study demonstrates why context matters: the same broad category, such as a provider-related barrier, may include incomplete follow-up, long waits, confusing explanations, or difficulty completing blood tests.

How can researchers compare treated and untreated patient segments?

Researchers can compare treated and untreated segments by applying the same analytic framework to both groups and then examining which themes differ in frequency, meaning, or sequence.

The November 20, 2026 study found that untreated participants described cost, language, transportation, information gaps, and healthcare navigation, while treated participants described homelessness, incarceration, addiction, and specialist delays as disruptions to care. Treated participants also described family support, provider trust, insurance coverage, treatment benefits, and stigma avoidance as facilitators.

How should healthcare teams interpret AI-generated themes?

Healthcare teams should treat AI-generated themes as an analytic starting point that requires researcher validation, not as a clinical diagnosis or an automatic policy recommendation.

The study authors reported limitations including the small qualitative sample, uneven representation of treated and untreated participants, and the absence of member-checking or participant transcript validation. Those limitations should remain visible in any AI-assisted synthesis.

How Evidano Supports AI Qualitative Analysis in Healthcare

Problem: Healthcare teams spend too long organizing transcripts

Evidano is an AI-powered qualitative data analysis platform that helps researchers analyze healthcare interviews, open-ended survey responses, and research documents.

Evidano can ingest interview transcripts and reports, then support thematic, content, frequency, and cross-segment analyses. Researchers studying HCV access could organize findings by treatment status, language, insurance, or other approved participant segments.

Teams can review related concepts through hierarchical codes and subcodes, word clouds, and co-occurrence networks. See the Evidano features overview for the available analysis workflows.

Problem: Multilingual interviews can hide important differences

Evidano supports transcription and translation workflows that can help research teams bring English- and Spanish-language interview materials into a consistent analysis process.

The November 20, 2026 PLoS One study shows why multilingual analysis requires care: Spanish-speaking participants described language barriers and a strong need for provider guidance. Translation should preserve meaning and researchers should review culturally sensitive interpretations before reporting them.

Problem: Researchers need to interrogate evidence without losing source context

Evidano provides AI chat over uploaded documents and analyses, allowing researchers to ask focused questions about themes, segments, and supporting passages.

A public health team could ask which barriers occur among untreated participants, which facilitators appear among treated participants, or where language and provider communication overlap. Researchers should verify important findings against the underlying transcript excerpts and study design.

Problem: Sensitive healthcare data requires strong safeguards

Evidano states that uploaded data is encrypted and is never used to train third-party models. Research teams should still apply informed-consent requirements, de-identification procedures, access controls, and institutional review policies.

The November 20, 2026 PLoS One study de-identified interview data by removing identifying information and assigning participant IDs. AI-assisted analysis should follow the same privacy-first principle.

FAQ: AI Qualitative Analysis in Healthcare

What is AI qualitative analysis in healthcare?

AI qualitative analysis in healthcare uses artificial intelligence to organize, code, compare, and summarize text such as patient interviews, open-ended surveys, clinical narratives, and research documents.

The method can accelerate thematic and cross-segment analysis, but the November 20, 2026 PLoS One study shows that human researchers remain responsible for interpretation, methodological transparency, and ethical oversight.

What were the main Hepatitis C treatment barriers in Los Angeles County?

The main barriers reported in the November 20, 2026 PLoS One study were personal circumstances, provider-related problems, insurance barriers, and inadequate access to medical care.

Interview themes added homelessness, incarceration, drug addiction, language barriers, cost, transportation, unclear information, long waits, and difficulty navigating the healthcare system.

How many people and interviews were included in the HCV access study?

The November 20, 2026 PLoS One study analyzed program data from 344 case-residents and conducted interviews with 10 participants.

The interview group included 7 untreated participants and 3 treated participants, while the program data included 168 treated and 176 untreated case-residents.

What facilitators helped patients access HCV treatment?

The November 20, 2026 PLoS One study identified motivation, family and community support, provider trust, adequate insurance, understanding treatment benefits, and the desire to avoid stigmatization as facilitators.

One treated participant summarized the role of provider trust by saying, "If they say this is a good idea, then I’m gonna do it."

Can AI replace qualitative healthcare researchers?

AI cannot replace qualitative healthcare researchers because researchers must assess context, bias, participant protection, sampling limits, and the meaning of competing interpretations.

The November 20, 2026 PLoS One study used collaborative coding and discrepancy resolution, practices that illustrate why accountable human review remains important even when software accelerates analysis.

Conclusion and Next Steps for Healthcare Research Teams

The November 20, 2026 PLoS One study shows that HCV treatment access is shaped by both measurable barriers and lived experiences involving language, cost, trust, housing, and care coordination. AI qualitative analysis in healthcare can help teams connect those layers by accelerating transcript review, thematic coding, segment comparison, and evidence retrieval. Researchers should use AI to extend analytic capacity while preserving human validation, participant privacy, and transparent reporting of limitations. Healthcare and public health teams can try Evidano for free to begin analyzing qualitative data more efficiently.

Topics

  • AI qualitative analysis in healthcare
  • AI thematic analysis healthcare
  • qualitative analysis of healthcare interviews
  • Hepatitis C treatment barriers
  • mixed-methods healthcare research

Keep reading

Browse all articles