A cross-sectional study was conducted from February 2025 to May 2025, using the quantitative survey-based method to assess the willingness, preferences, and feasibility of mHealth interventions for HIV prevention among the key populations in Pakistan, with data collected through collaboration with community-based organizations (CBOs) and nongovernment organizations (NGOs). The STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guideline (Checklist 1) was followed to conduct this study.

The minimum sample size for estimating the proportion willing to use mHealth was first calculated with the single-proportion formula (95% CI, 5% precision, and anticipated prevalence=50%), giving 384 individuals. Because the accessible sampling frame in our sites was approximately 500 individuals, we applied a finite-population correction [20], resulting in 218 individuals. Anticipating minor design clustering across the 3 key population strata, a design effect of 1.1 was used, yielding approximately 240 individuals. However, the final achieved sample size was limited to 210 due to practical constraints such as limited recruitment time and resource limitations. Furthermore, many individuals from marginalized populations were reluctant to participate because of fears related to stigma, confidentiality, and social consequences. Although the final sample size was slightly lower than the initial target, it does not meaningfully compromise the overall interpretability of our findings. A post hoc power check indicates that with a sample of 210 participants, the study still retains acceptable precision for estimating willingness proportions with approximately ±6% to 7% precision at a 95% CI [21,22].

The ethical approval for this study was obtained from the research and ethics committee of Riphah Institute of Pharmaceutical Sciences, Riphah International University (reference: REC-RIPS/RARE/2025/16); all ethical guidelines for research involving human participants were followed. Before participating, all participants were verbally informed about the purpose of the study, the nature of their involvement, potential risks and benefits, and their rights as research participants. Informed written consent was secured from all participants before data collection. Moreover, all the participants were informed that participation in this study was voluntary and that they could withdraw at any time. All participants were assured that their data would remain confidential, and they were informed that their data would be used for publication purposes. Participants were invited to self-identify their gender using inclusive options, with a choice to opt out, ensuring respect for individual autonomy and alignment with ethical standards for gender identity data collection. All the study procedures adopted complied with the Principles of the Declaration of Helsinki, Good Clinical Practices, and the applicable laws and regulations of research involving human subjects in Pakistan. The SAGER (Sex and Gender Equity in Research) guidelines were followed throughout the study and can be found in Checklist 2. Participants received PKR 500 (US $2.77) for study participation and an additional PKR 200 (US $1.11) for each peer successfully recruited to the study, up to a maximum of 5 peers.

We calculated the descriptive statistics for categorical variables, such as frequency and percentage, and mean and SD for the continuous variables. Cross-tabulation and chi-square tests were used to measure the association between the categorical variables. Bivariate and multivariate logistic regression analyses were conducted to identify the correlates of mHealth acceptance. Results were reported as odds ratio (OR) with 95% CI. Analyses were performed in SPSS (Statistical Package for the Social Sciences; version 27; IBM Corp) and RStudio (2025.05.0+496), an integrated development environment for R (version 4.5.0; R Foundation for Statistical Computing).

A total of 210 participants were included in the study (mean age of 29.55, SD 4.82 y). Of the 210 participants, 83 (39.5%) were people who inject drugs, 72 (34.3%) were transgender individuals, and 55 (26.2%) were men who have sex with men. Regarding sexual orientation, 84 (40%) identified as heterosexual, while 72 (34.3%) identified as homosexual. Most respondents were college graduates (n=88, 41.9%), while only a few reported having completed higher education (n=25, 11.9%). Employment status showed that half of the participants were unemployed (n=105, 50%), and a large proportion identified themselves as Muslims (n=159, 75.7%; Table 1).

Overall, 169 out of 210 (80.5%) participants reported having some knowledge of HIV and its transmission, while 29 out of 210 (17.1%) demonstrated a high level of HIV-related knowledge. Men who have sex with men reported the highest proportion of strong HIV knowledge (23/55, 41.8%), followed by transgender individuals (18/72, 25%) and people who inject drugs (13/83, 15.6%).

Of the 210 participants, 161 (76.7%) had smartphones, and 115 (54.8%) had daily access to the internet. The most common activities that participants used the internet on their smartphones for were online social networking apps like Facebook, WhatsApp, or TikTok; on average, they spent 3.41 (SD 3.10) hours per day on them. A total of 77 out of 210 (36.7%) participants used mobile apps several times a week, and 74 out of 210 (35.2%) used them daily. Additionally, 127 out of 210 (60.5%) of the participants stated that they had used an mHealth app, and 94 out of 210 (44.7%) were satisfied with their experience with the mHealth app. A significant association was found between the education levels and previous mHealth app experiences (χ²₄=24.08; P<.001). Among those who reported using the app (127/210, 60.5%), 40 out of 127 (31.4%) were transgender individuals, 27 out of 127 (21.2%) were men who have sex with men, and 32 out of 127 (25.1%) had a high school level education, suggesting a positive association between education and mHealth app engagement.

Of the 210 participants, 182 (86.7%) expressed a willingness to use a clinic-integrated mHealth app for HIV prevention (Table 2). More than half (n=108, 51.5%) indicated interest in a daily medication reminder feature, while 83 (39.5%) reported no interest in this function. Among those interested in medication reminders, 89 of 130 (68.5%) preferred to receive notifications through the app.

Regarding the dissemination of HIV prevention information, 110 out of 210 (52.4%) respondents preferred receiving updates on a weekly basis, followed by 49 out of 210 (23.3%) who favored monthly updates. Among these, 77 out of 180 (42.7%) indicated a preference for receiving such information via the app. Additionally, two-thirds of the participants (158/210, 75.2%) expressed interest in e-consultation services, with preferences split between app-based consultations (75/156, 48.1%) and phone calls (68/156, 43.6%). Furthermore, 134 out of 210 (63.8%) participants showed interest in receiving mental health information, with 73 out of 138 (52.9%) preferring to access this content through the app. Most participants (85/210, 40.5%) indicated that educational resources was the most important feature, followed by information on HIV self-testing (42/210, 20.2%) and secure messaging with health care providers (28/210, 13.5%). Furthermore, a total of 75 out of 210 (35.7%) participants reported moderate feasibility, and 61 out of 210 (29%) reported high feasibility for an mHealth app for HIV prevention.

There was a positive association between smartphone ownership, daily internet access, and interest in mHealth for HIV prevention; 99.1% (114/115) of them were willing to use an mHealth app for HIV prevention. A chi-square test indicated a statistically significant association between smartphone use, daily internet access, and willingness to use an mHealth app (χ²₂=71.19; P<.001). The likelihood-ratio chi-square test confirmed this association (χ²₂=63.35; P<.001). The Cramer V was 0.41, indicating a moderately strong association between internet access, smartphone ownership, and willingness to use a clinic-integrated mHealth app.

Out of 210 participants, 141 (67.1%) participants had been tested for HIV, but only 111 of 208 (52.9%) respondents knew their current HIV status. HIV prevention services uptake was limited, with 105 (50%) reporting no use; PrEP (47/88, 53.4%) was the most common. Self-reported awareness of STI status was moderate, with 92 out of 210 (43.8%) reporting that they knew their previous STI status. Participants reported a substantial level of vulnerability: 87 out of 210 (41.4%) had experienced violence, 128 out of 210 (61%) had been detained by the police, and 122 out of 210 (58.1%) used injectable drugs. Risky sexual behaviors were frequent, 104 out of 210 (49.5%) engaged in transactional sex, and 101 out of 210 (48.1%) reported multiple partners, while consistent condom use remained low (90/210, 42.9%), though slightly higher among men who have sex with men and transgender individuals (Table 3). Cross-tabulation analysis showed differences in patterns across groups. PrEP users reported higher use of a health-related mobile app. When grouped by sexual orientation, bisexual participants reported the highest levels of risky sexual behavior, followed by homosexual participants, while heterosexual participants reported the lowest levels. Participants who used any HIV prevention service tended to report moderately risky sexual behaviors, whereas those who did not use prevention services were more likely to report the highest levels of risk.

Part-time workers and students were significantly less likely to express a willingness to adopt a clinic-integrated mHealth app for HIV prevention compared to unemployed participants. Specifically, part-time workers had 70% lower odds (OR 0.30, 95% CI 0.12‐0.77; P=.01), and students had 87% lower odds (OR 0.13, 95% CI 0.02‐0.98; P=.047) of being willing to use the app. In contrast, participants who owned a smartphone were over 4 times more likely to be willing to adopt the mHealth app (OR 4.14, 95% CI 1.41‐12.21; P=.009). Moreover, individuals with prior experience using a mHealth app had nearly 3 times greater odds of willingness to adopt the app (OR 2.97, 95% CI 1.00‐8.81; P=.05) compared to those with no prior mHealth exposure (Table 4).

This study is the first in Pakistan to explore the feasibility and acceptability of mHealth interventions for HIV prevention among key populations, including men who have sex with men, transgender individuals, and people who inject drugs. The findings reveal a significant unmet need for accessible HIV prevention services, evidenced by high-risk behaviors and low engagement with existing methods. However, the widespread use of smartphones and social networking platforms presents a promising avenue for digital health solutions. Participants showed strong interest in a clinic-integrated mHealth app, favoring features like educational content, medication reminders, and e-consultation services.

Our findings identify several key features that prospective users desire in an mHealth intervention for HIV prevention. Specifically, participants expressed strong interest in receiving daily medication reminders, weekly prevention information, access to e-consultation with health care providers, and mental health resources. These preferences align with evidence from India and Nepal, where men who have sex with men and transgender individuals similarly valued medication and appointment reminders, HIV prevention information, and teleconsultation services. In the Nepali study, men who have sex with men also emphasized additional needs such as screening and monitoring illicit drug use, chemsex behavior, and ordering HIV-related supplies, reflecting broader expectations for comprehensive digital support. However, one notable difference is that most participants in Nepal preferred monthly messages and reminders, which contradicts our finding of strong interest in daily reminders and weekly prevention content. Preferences reported in India were more consistent with our results, as participants supported daily medication reminders and weekly HIV prevention information [1,29,30]. In our study, consistent condom use in the past 6 months was reported by 63.6% of men who have sex with men and 62.5% of transgender participants. These rates are considerably higher than those reported in Malaysia, where only 19% of men who have sex with men reported condom use, and moderately higher than in Bangladesh, where 48.6% of men who have sex with men reported consistent use [27,31]. However, these findings were comparable with those of Indonesia, where 76.97% reported protected sex in the past 6 months [32]. Limited access to HIV prevention information, weak social and health care systems, and criminalization of same-sex behaviors and cultural taboos in Muslim-majority settings such as Malaysia, Indonesia, and Bangladesh intensify privacy concerns and restrict access to condoms and prevention strategies. These constraints likely contribute to inconsistent condom use and reduced engagement with HIV prevention programs in these contexts [31,33,34]. mHealth programs have shown promise in improving condom use and mental health. In one example, a self-paced web-based intervention for Chinese men who have sex with men dating app users significantly reduced condomless anal sex and increased participants’ use of self-efficacy and positive attitudes towards condoms [35]. More broadly, digital interventions in the settings of low- and middle-income countries have generally been found to enhance sexual and reproductive health knowledge and attitudes, which support safer behavior [36]. Our study found that participants who reported using HIV prevention services such as PrEP were often involved in moderately high-risk activities, and this finding aligns with existing research showing that PrEP use can sometimes be associated with an increase in risky sexual behaviors and the incidence of bacterial STIs [37,38]. This may reflect risk compensation or that individuals with higher baseline risk are more likely to seek PrEP. However, this association is not universal; studies in South-East Asia showed that PrEP use does not always translate into risk compensation, and educational awareness may lead to less involvement in high-risk behavior [39,40]. These findings suggest that integrating HIV prevention, reproductive health information, and risk-reduction counseling into mHealth platforms may strengthen consistent condom use and reduce risky sexual practices.

Also, high rates of violence (41.4%) and police detention (61%) were observed among men who have sex with men and people who inject drugs. Research in the Middle East and North Africa, including countries such as Pakistan, Iran, Lebanon, and Ghana, highlighted a high level of violence, stigma, and discrimination among sexual and gender minorities, with police often implicated in discriminatory practices [41]. These legal and social pressures create significant barriers to accessing care, a finding that is consistent with qualitative research on the lived experiences of people with HIV in Pakistan who face pervasive stigma and structural challenges [13]. Studies from Malaysia and Iran reported that high stigma and criminalization towards men who have sex with men and transgender individuals increased the privacy concerns and limited the open health communication and emphasized mHealth as a discreet option [6,42]. Similarly, a cross-sectional survey conducted among Malaysian men who have sex with men reported that although 74.9% of participants expressed willingness to use a hypothetical HIV prevention app, concerns related to privacy and confidentiality significantly limited acceptability [43]. A Malaysian focus group study among men who have sex with men emphasized that encrypted messages, discreet notifications, and neutral app design are crucial to avoid accidental disclosure of HIV-related content [44].

HIV testing is the first step in reaching those at increased risk of HIV acquisition. Our study found that 63.7% of respondents rarely undergo HIV testing, highlighting the need for frequent screening. This is especially critical in Pakistan, where documented HIV outbreaks have been linked to gaps in testing and prevention among key populations [14]. mHealth could play a vital role in assessing risks and reminding and potentially motivating the marginalized community to be tested regularly through interactive quizzes and tailored push notifications [45,46]. A randomized control trial in Vietnam showed that mHealth improved knowledge and PrEP adherence through interactive quizzes and tailored push notifications [47]. As such, innovative technology can accelerate engagement in HIV prevention strategies and facilitate the PrEP uptake and HIV self-testing, especially in areas where these services are underused [48]. Our results highlight that there is considerable interest in specific mHealth strategies, such as receiving information related to HIV prevention, e-consultation, medication reminders, and mental health information. Given the lack of public dialogue and stigma, it is not surprising that the Pakistani marginalized population, like their counterparts [49,50], particularly China [51], Vietnam [52], and Malaysia [27], expressed interest. Our study reported low to moderate HIV knowledge, and most of the participants were unaware of their HIV status and STIs, which is confirmed by another study that reported a lack of awareness and low literacy rate among people living with HIV [53], highlighting the need for effective interventions to boost awareness and education. Several studies have reported that mHealth interventions can be helpful in enhancing HIV awareness and education, such as studies from India, Nepal, and Malaysia, demonstrated greater improvement in HIV knowledge awareness, HIV testing, PrEP uptake, and behavioral outcomes among marginalized communities [18,54,55]. Similarly, a cohort study in Indonesia showed that a mobile app increased HIV knowledge from 20% to 60% among men who have sex with men, from 22% to 57% among transgender individuals, and from 49% to 74% among people who inject drugs. Additionally, HIV testing uptake rose by 31% among men who have sex with men, 49% among transgender individuals, and 26% among people who inject drugs [50]. Our study found high feasibility for a clinic-integrated mHealth app that aligns with a systematic review that found that mHealth was both feasible and acceptable in reducing HIV incidence and enhancing care engagement in gay, bisexual, and other men who have sex with men communities [56]. 60.5% of the respondents reported that they had used clinic-integrated mHealth, and some of them had experienced using the smartphone for searching health-related information, which reflects the digital revolution that is especially explosive in Asia, and the fast-paced growth of technology and use within the community [27]. However, it is crucial to contextualize these findings regarding people who inject drugs, as our sample was recruited from rehabilitation centers. Among people who inject drugs in rehabilitation and treatment programs, mHealth interventions show high feasibility and willingness, with around 97% owning mobile phones and strong engagement in text-based HIV prevention and adherence support [57]. In contrast, people who inject drugs outside treatment face major barriers, including low phone ownership, unstable housing, device loss or sale, and limited literacy, reducing mHealth uptake [58]. For example, in Tanzania, only 32% of people who inject drugs owned phones, and just 6.4% had smartphones [59]. Evidence from Malaysia similarly shows higher mobile phone and retention among people who inject drugs in harm reduction programs compared to those outside care [60]. These findings indicate that mHealth works well for stabilized people who inject drugs in treatment but requires tailored strategies to reach broader community populations. Rapid advancements in mobile technology and app development are creating new opportunities to incorporate mHealth into existing HIV prevention services in the region. A mHealth-based app could serve as an additional tool to assist marginalized communities with HIV prevention or care needs between clinic visits, guiding them to needed services, potentially enhancing clinical care, and supporting them through screening and recommendations, and providing different modes of accessing information and prevention commodities. However, the high willingness to use reported in our study may not always translate into sustained use. Studies showed that real-world factors such as the affordability of mobile data, variable digital literacy, especially among marginalized populations, strongly impact users’ ability to navigate and benefit from these apps, with digital health literacy often requiring skills beyond general health literacy [61,62]. Mistrust of technology-based health services, rooted in historical health care disparities and concerns about privacy, further reduces long-term acceptability and use [63]. Addressing these barriers through community health worker support, culturally sensitive design, and policies promoting digital literacy and affordable connectivity is critical to translating initial interest into sustained mHealth app use.

Despite valuable insights, this study has several limitations. First, the sample size was limited to 210 participants due to resource constraints and challenges in accessing the hard-to-reach population. A larger, more diverse sample, especially including men who have sex with men and people who inject drugs from underdeveloped areas, could improve generalizability. Moreover, interpretation of the regression analysis should be approached with caution, as several predictors, including students, HIV knowledge, and prior mHealth experience, showed wide confidence intervals likely due to small subgroup sizes. Second, key populations like sex workers and prisoners were excluded due to legal and ethical constraints, notably the criminalization of sex work under Islamic and national laws. Third, reliance on self-reported data for HIV or STI status, sexual behavior, and drug use may introduce social desirability bias, especially given the criminalization of same-sex behavior and drug abuse under Section 377 [16] and the Narcotic Control Act [64]. While self-administered, anonymous questionnaires helped reduce this bias, future studies could ensure privacy using branching logic and adaptive questioning. Fourth, because RDS weights were not applied, estimates for men who have sex with men and transgender participants reflect sample characteristics rather than population-level estimates. This may limit generalizability and affect direct comparability with the convenience sample of people who inject drugs. Furthermore, recruiting people who inject drugs via convenience sampling from rehabilitation centers may overestimate mHealth feasibility compared to the broader, nontreated populations like people who inject drugs, who likely face greater structural and digital access barriers. Findings should therefore be interpreted with these sampling differences in mind. Finally, while participants expressed a high willingness to use mHealth, stated interest may not always translate to actual usage. Future implementation studies are needed to assess real-world uptake. Although this study focused on HIV-negative individuals, mHealth also shows potential for people living with HIV, particularly in supporting ART adherence and care management [6,26,65]. Future qualitative studies involving both people living with HIV and health care providers are recommended.

This study aimed to assess the feasibility, willingness, and preferences for mHealth interventions among men who have sex with men, transgender individuals, and people who inject drugs in Pakistan. Our findings demonstrate high acceptability and readiness to engage with a future mHealth solution tailored to their needs. Such platforms, yet to be developed, could overcome key barriers, including stigma, low HIV knowledge, and irregular testing, by offering discreet, accessible, and user-centered services. These should include features like HIV education, medication reminders, e-consultation, and mental health support. Policymakers, in collaboration with CBOs, should develop encrypted, culturally sensitive platforms with e-consultation and mental health support aimed at enhancing reach and mitigating HIV risk. Future research, including qualitative studies, is needed to optimize app design for these populations and people living with HIV.