Evidano is an AI-powered qualitative data analysis platform that helps researchers analyze interviews, open-ended surveys, and documents. According to the PLOS ONE article by Connell et al. (published July 30, 2026) PLOS ONE, help-seeking for mental health and substance use among people in contact with the criminal justice system is driven by the interaction of desire, ability, and local context. This post explains the study's methods and findings, highlights two concrete statistics from the paper (interviews with 50 people conducted between November 1, 2023 and July 31, 2024; sample split 23 in LA1 and 27 in LA2), and describes how AI-enabled qualitative research can accelerate and deepen this kind of relational analysis for researchers and service evaluators.
Key Takeaways
The PLOS ONE study by Connell et al. (published July 30, 2026) PLOS ONE shows that help-seeking for mental health and substance use among people under community justice supervision is not purely individual, it is shaped by social networks and local context.
- Connell et al. interviewed 50 people between November 1, 2023 and July 31, 2024, with 23 participants in LA1 and 27 in LA2, confirming relational influences across networks.
- Connell et al. report that nearly all participants self-identified as having current mental health or substance use needs and most had at least one unsuccessful help-seeking attempt as of July 30, 2026.
- Connell et al. identify three interacting domains that determine help-seeking: desire to seek help, ability to seek help, and help-seeking context; interventions need to operate at individual, network, and community levels.
- Participant testimony in Connell et al. includes stark phrasing such as "you just kind of can’t get away from the social" (Participant P13, Connell et al., 2026), illustrating network contagion and normalization effects.
What Happened: Study design and core findings
Answer: The PLOS ONE study conducted in-depth qualitative interviews to map how social networks influence help-seeking among people in contact with the criminal justice system living in the community.
According to Connell et al. in PLOS ONE (published July 30, 2026), the researchers applied framework analysis to interview data collected from November 1, 2023 to July 31, 2024 and combined that with participant-aided social network diagrams using Network Canvas.
Connell et al. (PLOS ONE, 2026) recruited 50 participants (23 in the largely rural LA1 and 27 in the urban, deprived LA2), transcribed interviews verbatim, and coded data into three main themes: desire, ability, and context.
Connell et al. report that relational influences appear in both the "desire to seek help" and the "ability to seek help, " and that network-level mistrust and stigma frequently inhibited access to services.
Direct participant quotes in Connell et al. illustrate these dynamics, for example, "I have dealt with it for almost twenty odd years by myself… I don’t like going to doctors because I think they are a waste of time" (Participant P12, Connell et al., PLOS ONE, 2026).
Findings Snapshot
| Date / Period | Metric | Value (from Connell et al., PLOS ONE) | Implication |
|---|---|---|---|
| Nov 1, 2023–Jul 31, 2024 | Interviews conducted | 50 participant interviews | Provides qualitative depth across two contrasting local authorities |
| Published Jul 30, 2026 | Geographic split | 23 participants in LA1, 27 participants in LA2 | Enables comparison of rural/affluent vs urban/deprived contexts |
| As reported Jul 30, 2026 | Consent for verbatim quotation | 49 of 50 participants consented to verbatim quotation | Permits rich, attributable qualitative excerpts for analysis |
Implications for qualitative researchers and service evaluators
Answer: Researchers should design qualitative studies that capture network-level data, not just individual narratives, because Connell et al. (PLOS ONE, 2026) show relational influences are central to help-seeking.
Connell et al. recommend multi-level interventions that address individual attitudes, network norms, and community-level service experience; qualitative evaluation should therefore code at the person, tie, and context levels.
Connell et al. (PLOS ONE, 2026) found that network mistrust and shared norms can normalize substance use and deter help-seeking, so evaluators need methods that detect contagion, homophily, and resource flows within networks.
For program design, Connell et al. suggest trauma-informed, relational practices and inclusion health approaches to improve service acceptability, particularly where networks share epistemic mistrust of statutory services.
How Evidano Helps: AI tools mapped to the study’s research problems
Problem: Large, verbatim transcripts slow synthesis
Answer: Evidano accelerates synthesis by automatically extracting themes, codes, and co-occurrence patterns from transcripts.
Evidano is an AI-powered qualitative data analysis platform that helps researchers analyze interviews, open-ended surveys, and documents, enabling thematic, content frequency, and cross-segment analyses that map individual and network-level themes efficiently.
Example use: ingest the 50 interview transcripts described in Connell et al. (PLOS ONE, 2026) and produce a matrix of desire, ability, and context codes with supporting quotations for each participant and network member.
Problem: Mapping relational influences across networks
Answer: Evidano supports network-aware qualitative coding and visualizations that surface relational patterns contributing to help-seeking.
Evidano can link coded segments to named network ties and generate co-occurrence networks and hierarchical code→subcode views so researchers can trace how a belief in one participant propagates across their network.
Use the Evidano features page to match platform visualizations to the social network analytic needs highlighted by Connell et al. (PLOS ONE, 2026).
Problem: Verbatim quotes and sensitive data management
Answer: Evidano provides transcription with PII redaction and encrypted storage to protect sensitive verbatim material.
Evidano’s secure workflows allow teams to retain pseudonymised quotations (as done in the Connell et al. dataset deposit) while controlling access in line with ethical approvals; see our data security information for details.
FAQ: qualitative analysis help-seeking
What is the main contribution of the Connell et al. PLOS ONE (2026) study to help-seeking research?
Answer: The main contribution is demonstrating that help-seeking is shaped by relational and contextual influences as well as individual factors.
Connell et al. (PLOS ONE, published July 30, 2026) used 50 qualitative interviews and social network diagrams to show that desire, ability, and context interact across social networks to enable or block help-seeking.
How should qualitative teams code for network effects in help-seeking studies?
Answer: Teams should code at multiple levels: individual beliefs and behaviors, tie-level influences (who says what), and community-level context.
Connell et al. (PLOS ONE, 2026) coded transcripts into desire, ability, and context themes and annotated network diagrams, a method that qualitative teams can replicate using software that links codes to named network members.
Can AI tools reproduce the depth of participant quotes used by Connell et al.?
Answer: Yes, AI tools can assist with extracting and organizing verbatim quotes while leaving interpretive synthesis to researchers.
Connell et al. relied on verbatim excerpts (49 of 50 participants consented to quotation) to illustrate themes; AI-enabled platforms like Evidano can pull and group such excerpts around codes to speed analytic iteration while preserving nuance.
Conclusion & Next Steps
Answer: Connell et al. (PLOS ONE, July 30, 2026) show that to understand help-seeking in community justice populations researchers must measure relational and contextual forces as well as individual factors.
Evidano helps teams operationalize that recommendation by linking transcript-level coding to network-aware visualizations and secure quote management, reducing manual synthesis time and improving transparency.
If you run qualitative or mixed-methods studies on help-seeking, consider ingesting your transcripts and network data into Evidano to generate thematic matrices and tie-level extracts that mirror the framework used by Connell et al.
Start a trial and see how AI-assisted thematic and network analyses speed interpretation and reporting: Try Evidano for free.