Qualitative analysis of cannabis and sexual behaviors is essential for clinics and researchers designing youth-centered harm reduction after Thailand’s 2022 cannabis policy changes. A mixed-methods study published June 26, 2026 assessed 200 young adults (mean age 21.45) attending key-population-led sexual health clinics in Bangkok and found past-month cannabis use = 22% and links to substance- or alcohol-influenced sex (PR=1.24 and PR=1.12 respectively). Read the original paper at PLOS ONE. In this post we distill what matters for UX researchers, clinic leads, and policy teams and show a reproducible workflow to turn transcripts, surveys, and interview audio into thematic, segment, and cross-tab analyses.
Key Takeaways
This post summarizes a June 26, 2026 mixed-methods study (n=200) that links past-month cannabis use to higher self-reported alcohol- or substance-influenced sex but not to higher measured STI/HIV rates in this sample.
The study shows how to prioritize harm-reduction messages that respect user agency while addressing co-use with alcohol as a likely mediator.
- Past-month cannabis use was 22% in the clinic-based sample of 18–24-year-olds (mean age 21.45).
- Cannabis use correlated with alcohol- or substance-influenced sex (PR = 1.24, 95% CI 1.09–1.41) and alcohol-influenced sex (PR = 1.12, 95% CI 1.02–1.24).
- PrEP use (34% current users) and HIV prevalence at enrollment (1.5%) did not differ significantly by cannabis use in adjusted analyses.
- When planning interventions, target polysubstance contexts: 61% of cannabis users reported co-using alcohol in this study.
Fast take + source
This June 26, 2026 mixed-methods study (n=200) combined a tablet survey, clinic lab data, and 30 semi-structured interviews to map cannabis patterns, sexual behaviors, and PrEP outcomes among 18–24-year-olds at two Bangkok clinics.
- Source: PLOS ONE (Yimsaard et al., published June 26, 2026).
- Why it matters: cannabis users were more likely to report alcohol- or substance-influenced sex, but not higher measured STI/HIV rates in this sample.
- Quick payoff: use thematic and segment analysis to design youth-centered messages that acknowledge agency while reducing intoxication-related risk.
Findings snapshot
| Metric | Value | Source / note |
|---|---|---|
| Sample | 200 young adults (18–24) | Clinic-based convenience sample (Nov 27, 2023–Mar 29, 2024) |
| Past-month cannabis use | 22% | Self-report; CUDIT-R used for severity |
| Mean age | 21.45 yrs (SD 1.79) | Survey demographics |
| Association: substance-influenced sex | PR = 1.24 (95% CI: 1.09–1.41) | Poisson regression |
| Association: alcohol-influenced sex | PR = 1.12 (95% CI: 1.02–1.24) | Poisson regression |
| PrEP users | 34% current users; 57% report very good/excellent adherence | No significant difference by cannabis use |
| HIV prevalence (enrollment) | 1.5% | Lab records; no significant difference by cannabis use |
What the study did and found
The study ran an anonymous tablet survey (n=200), extracted same-day HIV/STI test results from clinic records, and conducted 30 Thai-language semi-structured interviews (15 users, 15 non-users).
- Quantitative methods: chi-square tests and Poisson regression; qualitative methods: thematic coding in Thai (Dedoose).
- Key quantitative results: past-month cannabis use = 22%; cannabis correlated with alcohol- or substance-influenced sex but not with measured STI/HIV outcomes or PrEP adherence in adjusted models.
- Key qualitative themes: lowered inhibition and impaired judgement versus user narratives emphasizing routines and individual responsibility.
- Interpretation: cannabis often appears in polysubstance contexts (61% of cannabis users co-used alcohol), making alcohol co-use an important mediator to address.
So what for researchers and clinic teams
UX & behavioral researchers
UX and behavioral researchers should design interventions that reflect divergent user narratives, balancing routines that preserve adherence with concerns about impairment.
Design interventions that reflect divergent user narratives: some users report routines that preserve PrEP adherence; others and many non-users worry about impairment. Segment messaging by self-reported use patterns (e.g., occasional vs. co‑use with alcohol).
Use mixed-methods coding to link themes (e.g., 'lowered inhibition') to measurable outcomes (e.g., substance-influenced sex) when prioritizing features or content.
Clinic leads & public health teams
Clinic leads and public health teams should integrate brief substance screening into sexual health visits and triage patients for harm-reduction counselling focused on co-use and situational risk.
Integrate brief substance screening (AUDIT-C + CUDIT-R) into sexual health visits and triage patients for harm-reduction counselling focused on co-use and situational risk (bars, parties).
Avoid blanket risk messaging; co-design non-stigmatizing scripts that acknowledge agency and offer concrete tactics (carry condoms, time PrEP dosing before going out).
Policy & evaluation
Policy teams should monitor polysubstance trends after policy changes and combine lab data with behavioral context in surveillance.
Monitor polysubstance trends after policy changes: decriminalization can increase availability but not uniformly change infection rates, surveillance should combine lab data with behavior and context.
Prioritize controlled-access datasets for sensitive topics (authors note data cannot be public) and pre-register mixed-methods analysis plans to improve causal claims.
Do more, faster with Evidano
From Thai audio to themes in hours
Evidano is an AI-powered qualitative data analysis platform that speeds Thai audio transcription, PII redaction, and thematic coding so teams can start analysis sooner.
Transcribe interviews with PII redaction and a custom Thai dictionary, then auto-translate and preserve original transcripts for audit. Evidano reduces manual transcript prep so thematic coding starts sooner.
Triangulate survey, lab, and interview data
Evidano links survey, lab, and interview data in a single workspace to run cross-segment analyses and produce prevalence ratios and co-occurrence outputs.
Import survey CSVs and lab spreadsheets; link participant IDs (securely) and run cross-segment analyses (e.g., cannabis users who co-use alcohol vs. cannabis-only). Evidano outputs prevalence ratios, co-occurrence networks, and segment-based theme frequency in one workspace.
Codebook reuse and reproducible analysis
Evidano supports codebook reuse and AI-assisted coding to improve consistency and reproducibility across projects.
Upload existing codebooks or let Evidano propose hierarchical codes → subcodes. Apply AI-assisted coding for consistency and export clear quote lists for stakeholder briefs, ideal for clinic training or policy memos.
Security and compliance
Evidano encrypts data end-to-end and keeps proprietary models from training third-party systems, supporting sensitive sexual health work and controlled datasets.
Data is encrypted end-to-end and Evidano proprietary models are not used to train third-party systems, important when handling sensitive sexual health records and controlled datasets.
Checklist: run this study-to-action workflow in 10 days
This checklist lays out a 10-day run-book to translate similar mixed-methods work into interventions using transcript, survey, and lab data.
- Day 1–2: Import survey CSV + lab results into your analysis workspace; map IDs and variables.
- Day 3: Run frequency and cross-segment analyses (past-month use, co-use, PrEP status).
- Day 4–5: Upload interview audio, auto-transcribe with Thai dictionary and PII redaction.
- Day 6: Use AI to propose initial codebook; review and lock codes with clinicians.
- Day 7: Generate thematic reports, co-occurrence networks, and top quotes per segment.
- Day 8–9: Draft youth-centered harm reduction messages and clinic scripts using quote evidence.
- Day 10: Export stakeholder-ready slides and a one-page executive brief.
Ethics note
This post summarizes research findings for program design and evaluation and is non-diagnostic and for public health planning only.
The original authors restrict public release of individual-level data for confidentiality; controlled access requests are managed by IHRI as described in the paper.
Wrapping up: next moves
If you run mixed-methods studies on substance use and sexual health, prioritize analyses that link situational intoxication (alcohol plus cannabis) to behavior and build messages that respect users’ sense of agency.
- Ready to reproduce these outputs on your transcripts and surveys? See the original study at PLOS ONE and then Try Evidano for free.
- Next two moves: (1) run a 10-day pilot using the checklist above; (2) test two harm-reduction message variants (routine-first vs. risk-first) and measure uptake.
FAQ: Cannabis & sexual behaviors
What did the study find about cannabis and sexual behavior?
The study found that past-month cannabis use was associated with higher self-reported alcohol- or substance-influenced sex but not with higher measured STI/HIV rates in this sample.
Specifically, cannabis use correlated with substance-influenced sex (PR = 1.24, 95% CI 1.09–1.41) and alcohol-influenced sex (PR = 1.12, 95% CI 1.02–1.24).
Did cannabis use mean more STIs or HIV in this sample?
No, cannabis use did not correspond to higher measured STI or HIV rates in adjusted analyses in this sample.
The study reports HIV prevalence at enrollment of 1.5% and found no significant difference by cannabis use when comparing lab records.
How common was past-month cannabis use among participants?
Past-month cannabis use was 22% in the clinic-based convenience sample of 18–24-year-olds.
The sample mean age was 21.45 (SD 1.79) and the study ran from Nov 27, 2023 to Mar 29, 2024 for recruitment.
What should clinics do differently based on these findings?
Clinics should integrate brief substance screening and offer harm-reduction counselling that acknowledges agency and targets polysubstance situations like bars and parties.
Suggested actions include AUDIT-C + CUDIT-R screening, non-stigmatizing scripts, and concrete tactics such as carrying condoms and timing PrEP dosing before events.