Study design

This study employed a mixed-methods design, consisting of two distinct phases: quantitative followed by qualitative [11]. This study was approved by Chulalongkorn University Institute Review Board (COA No. 1262/2023). This manuscript was reported following the Good Reporting of a Mixed Methods Study (GRAMMS) guidelines [12].

Setting and sample

This study was conducted at two key-population-led community sexual health clinics: The Pribta-Tangerine clinic and the Rainbow Sky Association of Thailand clinic, both located in Bangkok, Thailand. Both clinics provide comprehensive sexual health services, including HIV testing, PrEP, and STI screening and treatment, to individuals of all genders, with a focus on key populations such as MSM and transgender individuals. The target sample was 200 young adults (aged 18–24 years) attending the clinics for either HIV prophylaxis, HIV testing, or STI testing, including syphilis, gonorrhea, chlamydia, hepatitis C (HCV), and hepatitis B (HBV). Exclusion criteria included known HIV-positive status or prior treatment for HIV before enrollment, as the study focused on HIV prevention and sexual behaviors among individuals at risk, whose contexts differ from those living with HIV.

The sample size was estimated using Cochran’s formula [13], based on a 2% prevalence of cannabis use in Thailand [14], with a precision of 2%. To account for an anticipated 5% attrition, the final sample size was set at 200 participants.

For the qualitative phase, a subset of 30 participants was purposively selected from the 200 participants in the quantitative phase, consisting of 15 cannabis users and 15 non-users.

Participants were offered an information sheet and provided written consent before participating in both quantitative and qualitative phases of the study.

Data collection and measures

Sexual behavior data and PrEP adherence

Sexual behavior data were collected using a study-specific questionnaire, where participants were asked whether they had engaged in vaginal or anal sex in the past 3 months. Those who answered “yes” were then asked to report the number of sexual partners they had for each type of sex, as well as the frequency of condom use during intercourse, with the following response options: every time, sometimes, rarely, and never. Additionally, participants were asked to rate how often, in the past 3 months, they had used any drugs or consumed alcohol before or during sex, and whether they had exchanged sex for money, using the same response options.

Participants were also asked whether they currently use PrEP. If they did, they were further asked to rate their PrEP adherence over the past month using a 6-point Likert scale, ranging from “very poor” to “excellent.”

Alcohol, tobacco and substance use data

Participants were asked whether they had ever consumed alcohol in their lifetime. Those who reported lifetime alcohol use were then asked to complete the Alcohol Use Disorder Identification Test-Consumption (AUDIT-C). An AUDIT-C score of ≥4 was considered indicative of unhealthy alcohol use in this study [15]. For tobacco use, participants were asked whether they currently smoke, with response options of not at all, daily, or less than daily. Other substance use was assessed by asking participants whether they had ever used any of the following substances: cannabis, amphetamine-type stimulants, cocaine, laughing gas, poppers, sleeping pills, hallucinogens, heroin, ecstasy, tramadol, kratom and other substances. Participants who reported lifetime use of any substance were subsequently asked whether they had used it in the past month.

Cannabis use data

Participants who reported past-month cannabis use were asked to complete the Cannabis Use Disorder Identification Test-Revised (CUDIT-R). The CUDIT-R is an 8-item self-report screening tool that assesses various aspects of cannabis use and related problems, including frequency, quantity, and negative consequences. The total score ranges from 0 to 32, with a score of 8 or higher indicating hazardous use and a score of 12 or higher indicating probable cannabis use disorder [16].

We also assessed whether participants have ever used cannabis simultaneously with any of the following substances: alcohol, amphetamine-type stimulants, laughing gas, poppers, sleeping pills, hallucinogens, heroin, ecstasy, tramadol, kratom and other substance. Participants responded with “yes” or “no” to each substance.

Participants were also asked to report their most frequent method of cannabis consumption in the past month, with options including inhalation (joint, blunt, bong, vaporizer, dabbing), ingestion (food, drink, capsule, CBD oil), sublingual spray, and topical application.

Laboratory Data

Laboratory data were extracted from clinic records based on routine testing conducted as part of each clinic’s standard clinical care. For all participants who completed the survey (n = 200), we retrieved HIV and other STI test results (gonorrhea, syphilis, chlamydia, HBV, and HCV) recorded on the date of survey completion. As testing was performed according to their clinical guidelines, not all participants were tested for all conditions.

Qualitative phase

During the qualitative phase (May 21, 2024 – September 19, 2024), we purposively selected 15 cannabis users and 15 non-users from the 200 survey respondents, resulting in a total of 30 participants. These participants were invited to take part in semi-structured interviews (SSIs) to explore perceptions of cannabis use in relation to sexual behaviors and HIV prevention behaviors. The interview guide was developed based on the study objectives and existing literature on cannabis use and sexual risk behaviors. Examples of questions used in the SSIs are provided in Supplementary Table 1.

The SSIs were conducted in private rooms at the clinics. All SSIs were conducted in Thai, audio recorded with participants’ consent, and ranged in duration from 32 to 62 minutes, with an average length of 46.07 minutes (SD = 9.16). Participants received the local currency equivalent of 24 USD as compensation for their time. Each SSI was facilitated using a structured interview guide by either the principal investigator (a psychiatrist), an MSc-level staff member, or a PhD-level staff member, all of whom were cisgender women trained in qualitative research.

Data analysis

Triangulation of data

By employing a triangulation approach, this study utilized convergence analysis to compare findings from different methods to expand understanding by examining the extent to which the quantitative and qualitative results converged or diverged. We first analyzed quantitative and qualitative results separately and then compared results. Where possible, we have an integrated presentation of results.

Quantitative survey

Sample characteristics.

At the enrollment, 200 participants were recruited. The average age of participants was 21.45 years (SD = 1.79). Eighty-five percent were assigned male at birth with 35.5% identifying themselves as gay men, 32% as transgender women, and 11% as men. The majority (70%) of participants reported having completed education beyond grade 12. Additionally, 57% of participants were students and 8% reported living with long-term partners (Table 1).

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Sociodemographic Characteristics, Substance Use, Sexual Behaviors, HIV Prevention Behaviors, and HIV/STI Laboratory Results Among Young Adults Attending Key Population-Led Sexual Health Clinics in Bangkok, Thailand: Overall and Stratified by Past-Month Cannabis Use (N = 200).

https://doi.org/10.1371/journal.pone.0352395.t001

Sexual behaviors and PrEP adherence

Almost sixty percent (59%, n = 118) of participants reported having engaged in anal sex, while 18% (n = 36) reported vaginal sex in the past 3 months. Among those, more than half reported inconsistent condom use during vaginal (63.9%) and anal sex (53.4%). Approximately half of the participants reported having more than one sexual partner, including 55.6% for vaginal intercourse and 47.5% for anal intercourse. Forty-three percent of participants engaged in sexual activity while consuming alcohol, 16.5% engaged in sexual activity while using drug, and 8.5% reported exchanging sex for money (Table 1). Chi-square analysis revealed that cannabis users were significantly more likely to report alcohol-influenced sex (70.5% vs 35.3%, χ² = 17.35, p < 0.001), substance-influenced sex (40.9% vs 9.6%, χ² = 24.35, p < 0.001) and engaging sex for money (20.5% vs 5.1%, Fisher’s Exact Test, p = 0.003). Around one-third (34%) of participants were current PrEP users. More than half (57.4%) of PrEP users rated their PrEP adherence over the past month as either “Very Good” or “Excellent”.

Alcohol, tobacco and substance use data

Most participants (93%) reported lifetime alcohol use, with 52.2% having an AUDIT-C score ≥ 4. Eleven percent were current smokers. Nearly half of the participants (48.5%) reported no substance use in the past month, while 30.5% reported using a single substance, and 21.0% reported using ≥2 substances. Cannabis was the most reported substance used in the past month (22%), followed by poppers (21%) and laughing gas (11%) (Table 1).

Cannabis-related data

Past-month cannabis users were more likely to have an AUDIT-C score ≥4 (75% vs. 45.1%, χ² = 12.06, p < 0.001) and to report daily smoking (13.6% vs. 3.8%, Fisher’s Exact Test, p < 0.001) compared to non-users. They were also significantly more likely to report using two or more substances (excluding alcohol and smoking) in the past month (68.2% vs. 7.7%, χ² = 87.19, p < 0.001) (Table 1).

Among past-month cannabis users, 11.4% were classified as possible cases of cannabis use disorder and 20.5% as hazardous users, based on the CUDIT-R. Smoking joints was the most reported method of consumption (38.6%). Nearly two-thirds (61.4%) reported co-using alcohol.

Laboratory Data

At enrollment, the HIV prevalence among our sample was 1.5% (n = 200). Among the participants tested, 7.6% had positive VDRL results (n = 185), 22.6% tested positive for gonorrhea (n = 106), 14.3% tested positive for Chlamydia (n = 105), and 2.6% tested positive for HCV (n = 114). There were no significant differences between cannabis users and non-users for any of these infections (Table 1).

Qualitative interview

Thirty SSIs participants included 15 cannabis users (Mean age 21.87, SD = 1.99, 40% transgender women, 46.7% student, 93.3% single) and 15 non-cannabis users (Mean age 20.87 years, SD = 2.03, 40% transgender women, 53.3% students, 100% single) (Table 2).

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Sociodemographic Characteristics of Young Adults Participating in Qualitative Interviews (N = 30).

https://doi.org/10.1371/journal.pone.0352395.t002

Sexual Behavior and Cannabis use

Quantitative analysis revealed that cannabis use was significantly associated with engaging in sex under the influence of any substance (Prevalence Ratio (PR) = 1.24, 95% CI: 1.09–1.41) and alcohol-influenced sex (PR = 1.12, 95% CI: 1.02–1.24) (Table 3). Complementing these findings, qualitative data revealed an emerging theme of lowered inhibition, characterized by a diminished capacity for self-control that led participants to engage in situations they might otherwise avoid, such as accepting substances or engaging in unplanned sex. As one participant shared:

Download:

• PNG
  larger image
• TIFF
  original image

Table 3. Associations Between Past-Month Cannabis Use and Sexual Behaviors Among Young Adults Attending Key Population-Led Sexual Health Clinics in Bangkok, Thailand (N = 200): Results from Poisson Regression Analysis.

https://doi.org/10.1371/journal.pone.0352395.t003

“I never got cocaine myself. I always received it from my partner. Normally, if it wasn’t available, I wouldn’t even think about using it. But when he had it and offered, I would just go along with him. However, on days when I wasn’t using cannabis, there were times when I was able to refuse cocaine.” [Cannabis user-01]

Another participant described how cannabis use contributed to spontaneous sexual encounter:

“I didn’t use weed before going out, but I used it during my night out, like when I went to a bar at Khaosan Road, and I ended up having sex. It was totally unplanned. I didn’t intend to have sex that day.” [Cannabis user-07]

While the quantitative data did not show a statistically significant relationship between cannabis use and inconsistent condom use, multiple sexual partners, HIV serology or other STI results (Table 1), the qualitative findings highlighted impaired judgement as an important contextual factor. Many participants described how cannabis intoxication might diminish their capacity to make safer sexual decision. One cannabis user explained:

“Once I use cannabis, I stop paying attention to things like making sure to wear a condom or checking for anything. Using cannabis, for me, is mostly not about sitting quietly and using it. The environment plays a role too. Everything blends, and you don’t really pay attention to whether there’s a condom, lubricant, or anything.” [Cannabis user-01]

Non-cannabis users, despite lacking direct experience with cannabis, also expressed concerns about its potential to impair judgement during sexual activity. As one participant remarked:

“I think using cannabis can make you feel spaced out and lose control, similar to being drunk. While I’ve never used cannabis, I believe its effects could impair judgment during sexual activity, much like alcohol, which can also lead to similar outcomes.” [Non user-08]

However, not all participants agreed. Three cannabis users insisted that cannabis use does not inherently increase sexual risk and instead emphasized individual responsibility:

“No, not at all. I’ve had experience using cannabis before having sex. At that time, it was completely normal—like, I told him to use a condom, and he used it as usual. I think risky behaviors like forgetting to use a condom or engaging in group sex depend entirely on the individual, whether they choose to do that or not.” [Cannabis user-05]

PrEP Adherence and cannabis use

Although the quantitative analysis did not find a significant difference in PrEP adherence between cannabis users and non-users, qualitative findings revealed additional insights into perceived differences between users and non-users.

One emerging theme was impaired judgement, primarily described by non-users when reflecting on how cannabis might affect PrEP adherence. Several non-users perceived cannabis as a substance that could interfere with one’s ability to remember or prioritize taking PrEP. One participant stated:

“For example, if someone consumes cannabis to the point of intoxication and needs to take PrEP at 8 p.m., they might be too impaired, causing them to forget their PrEP for that day. I believe it will happen 100% of the time.” [Non-user 01]

Another highlighted the role of dosage:

“I think it depends on how much a person consumes. If they use it to the point of feeling completely spaced out, it will definitely have an effect. They probably wouldn’t be conscious enough to think about what time they need to take their PrEP.” [Non-user 08]

Others noted that cannabis might distort how individuals perceive the importance of PrEP:

“I think if someone smokes cannabis, they might feel relaxed or in a daze, and they might think PreP is not a big deal. They may know they need to take PrEP, but they don’t feel it’s urgent. It’s not that they forget, but they just feel like it’s not that important.” [Non-user 02]

In contrast, several cannabis users described their own ability to maintain adherence as a matter of individual responsibility. Many users highlighted that they had developed a routine that allowed them to take PrEP before using cannabis. One participant stated:

“I think it depends on when a person takes PrEP. Since PrEP can be taken at any time of the day. For me, I take it in the evening, so it hasn’t been a problem. I usually only use cannabis when I go out at night.” [Cannabis user-01]