Assessing consistent condom use among migrant men who have sex with men in Shanghai, China: validation of an information–motivation–behavioural skills model

China is a country with vast territory and imbalanced development across different regions, and now it is experiencing massive internal migration of people in search of employment opportunities mostly from rural to urban areas, which is similar to cross-border migrations [1‐3]. Reportedly, there are about 253 million migrants in China, accounting for about 20% of the country’s total population [2, 4, 5]. Migrants are typically excluded from the general health care system in host cities, and researchers have found migrants are more likely to engage in risky sexual behaviours, which makes them more vulnerable to human immunodeficiency virus (HIV) infection as compared with residents [1, 4, 6, 7].

Recently, MSM have accounted for about one-third of China’s new HIV infections [8‐10]. Several studies have indicated more than half of MSM in China were migrants, with a proportion ranging from 55 to 88% [3, 11]. While they are part of the migrant population in search of better employment opportunities, MSM in China might have additional motivations to migrate, so they could hide their sexual orientation from mainstream heterosexual society in their home towns [12‐16]. This strong discrimination against homosexuality and internalized MSM related stigma might further limit their ability to access HIV prevention and care in their host cities [15, 17‐20]. They were unwilling to seek for professional help from medical professionals because they were reluctant to participate into public health activities due to homophobia and they didn’t want to disclose their sexual orientation to others [7, 15, 16]. The disclosure of their sexual identities might even pose potential threats to their housing or job security [7, 21]. In addition, migrants MSM only have limited access to health care not only because some of the health care benefits were only available to local residents due to Chinese official household registration policy, known as “Hukou”, but also because migrant MSM could only find job in informal labour markets without job security or health insurance [3, 4, 7, 12, 15]. More importantly, migrant MSM have higher odds of engaging in high risk sexual behaviours, such as reporting unprotected anal intercourse and having multiple sexual partners which further complicates the spread of HIV among them [1, 3, 4, 12, 13]. It has been reported that migrant MSM in China have greater likelihood of being infected with HIV compared to resident MSM [4, 7, 22‐25]. A study conducted in eight cities in Shandong Province reported that prevalence of HIV infection among migrant MSM ranged from 2.3 to 8.9%, which were higher than the 0.9 to 4.0% range found in resident MSM [1]. In addition, infected migrants may serve as a bridge population and facilitate HIV transmission when they move between rural and urban areas [1, 3, 13, 26]. The increasing prevalence of HIV infection among MSM in China owes mainly to inconsistent condom use during anal sex [7, 16, 24, 27, 28]. Some studies have indicated migrant MSM were considerably less likely to consistently use condoms as compared with local MSM [1, 4, 13, 29]. It is important to focus on interventions for promoting condom use among migrant MSM [7, 13, 27].

Factors associated with condom use among migrant MSM have differed in the literature [12, 22, 24, 25, 27‐31]. Some studies found that a lack of HIV-related knowledge was associated with high risk sexual behaviors, migrant MSM were at greater risk of HIV acquisition and it was important to carry out health education and enhance knowledge about HIV infection [4, 5, 15, 24‐26]. While social psychosocial and cultural factors were given increasing importance among this population, others studies indicated social connections and social support played a role in consistent condom use among migrants, and findings underscored urgent need for interventions focused on strong social support network for migrant MSM [1, 5, 7, 13, 14]. Migrants were often socially isolated from their families and friends [2]. On the one hand, this migration gave them a chance to escape from social norms and traditional values [7, 12]. City life meant less traditional sexual lifestyles and more exciting experience, it was relatively easier for them to find same sex partners in cities, and they might look for intimacy through sex, which might contribute to the increase of unprotected anal intercourse [4, 7, 15, 32]. On the other hand, it was widely reported that psychosocial problems were associated with high-risk sexual behaviors among MSM [7, 14, 19, 20]. While migrant MSM were socially marginalized and discriminated, they suffered from different psychosocial problems, which further complicated their risk sexual behaviors among them [7, 15, 19]. Some research pointed out that lack of perception of HIV risk may also relate to risky sexual behaviours. Statistics showed that although most MSM in China were aware of the high prevalence of HIV infection among this population, they still believed that they were unlikely to be infected with HIV [24, 31, 33, 34]. Some migrant MSM perceived that sex with regular partners were safe and they might not use condom during anal intercourse [15]. Factors associated with inconsistent condom use among migrant MSM are multidimensional, vary among different samples, and owe to complicated interactions between factors [35, 36]. Some recent studies have applied the information–motivation–behavioural (IMB) skills model to explain interactions between different HIV-related risk factors which was developed by Fisher and his colleagues [37]. The constructs of the IMB model are regarded as highly generalizable determinants of HIV-preventive behaviours in any population, such as gay men and heterosexual university students, senior high school students, unmarried rural-to-urban female migrants, male street workers, new sex partners met online or offline, HIV-positive gay and bisexual Men [37‐42]. Good fitness of the model has been shown among HIV high-risk populations in studies predicting condom use [43‐45]. The IMB model was intended to integrate biomedical and behavioural approaches in stemming the HIV epidemic [37, 39, 43, 46, 47]. Based on previous literature about migrant MSM in China, we found HIV-related knowledge, perceived risks of HIV infection, and social support were associated with risky HIV-related behaviours [1, 2, 4, 13, 24, 31, 32]. Seminal research on the IMB model suggested that attitudes toward condom use may affect motivation to prevent HIV transmission, and people with skills in negotiating safer sex and refusing unsafe sex may have less difficulty performing HIV-prevention behaviours [37, 46, 47]. However, to our knowledge, no study has applied the IMB model to assess consistent condom use among migrant MSM in China.

Assuming information (on sexual and non-sexual transmission of disease) and motivation (attitudes toward condom use, perceived risks, and social support) would affect consistent condom use through behavioural skills (condom use negotiation skills and unprotected sex refusal skills), while information and motivation may also directly affect condom use, we tested associations among IMB constructs as predictors of consistent condom use. To do this, we used structural equation modelling to study migrant MSM in Shanghai.

HIV prevalence among the study participants was 4.4%; slightly lower than results from other studies [4, 22, 29]. The figure may be underestimated because 36.0% of the participants reported unknown HIV testing results. Additionally, 44.9% reported having practiced consistent condom use in the preceding 6 months, which was close to figures found in other studies [4, 55]. Prompt action to prevent further HIV transmission among migrant MSM appears to be a vitally necessary step. The present study focused on the applicability of the IMB model in predicting consistent condom use among migrant MSM, and illustrated unique relationships among different constructs regarding consistent condom use in this population. Both the full and restricted models indicated behavioural skills directly affected condom use behaviour (β = 0.779; β = 0.629), and motivation directly affected behavioural skills (β = 0.645; β = 0.626), while information had no direct effect either on behavioural skills or on condom use behaviour. The model assumed that information and motivation affect consistent condom use via behavioural skills while information and motivation only have indirect effect. More motivation(i.e. risk perception, attitude toward using condom and social support), and more behavioural skills (i.e. condom negotiation skills and unprotected sex refusal skills)were associated with consistent condom use [38]. These findings were similar to the results from other studies [38, 40, 56]. It was important to improve their attitude toward condom use, provide them with more social support, train them about condom use skills and educate them about how to refuse unprotected sex, which might increase consistent condom use among migrant MSM.

Most intervention strategies thus far targeting MSM and including migrants have mainly focused on improving acquisition of HIV-related information [4, 24, 57, 58]. However, owing to the increasing presence of new forms of media, migrant MSM potentially have increased opportunities to access information on HIV prevention [15, 36, 55, 59]. As for information on sexual and non-sexual transmission disease respectively, most participants could answer the questions correctly, which meant that ceiling effects of knowledge acquisition may exist [60]. This may explain why most intervention methods emphasizing knowledge transferal have only had moderate success [10, 59, 61]. Additionally, the correlation between information and motivation was 0.553, while motivation had an indirect relationship with condom use behaviour through behavioural skills [40]. Therefore, other constructs apart from imparting knowledge should be included in HIV-prevention strategies, such as heightening motivation and behavioural skills. These aspects may also further encourage consistent condom use [38].

Motivation directly affected behavioural skills and indirectly contributed to condom use behaviour via behavioural skills. We should therefore raise individuals’ motivation with regard to safe sex behaviours. One possible method is to provide more social support for migrant MSM, seek to change their attitudes toward condom use, and help them to better perceive the potential dangers of high-risk sexual behaviours such as inconsistent condom use. As for social support, previous studies already indicated migrants were often socially isolated from their families and migrant MSM often suffered from lack of love and intimacy, and this may contribute to more prevalent sexual risks among them [1, 4, 5, 13, 14, 32]. As families in China have always played an important role in providing mutual support to their members and stronger social support network could equip them with better coping mechanisms [7, 10], we suggest possible inclusion of social support-based intervention in our intervention strategies [13]. By doing so, migrant MSM with greater social support may in turn experience greater motivation to practice safer sex. This type of social support may involve providing them emotional support when they are in need of it [50]. Regarding perceived risks and attitude toward condom use, migrant MSM have been found to typically lack awareness regarding risks of unsafe sex and HIV infection. Therefore, future intervention strategies should pay greater attention to heightening their awareness of their vulnerability, and fostering more positive attitudes toward condom use [24, 31].

Behavioural skills were found to contribute directly to consistent condom use, and were the most important examined factor affecting condom use behaviour. Migrant MSM should be provided with greater skills for improving their abilities to negotiate condom use and to refuse unprotected sex. This can be accomplished in ways such as encouraging them, and providing guidance on how, to discuss condom use with their partners, or to refuse sex if the partner refuses to use a condom [37, 51].

In short, neither information nor motivation were found to be directly associated with consistent condom use, while they might have indirect effects via behavioural skills. The participants also already generally possessed common knowledge about sexual and nonsexual transmission of the disease; therefore, knowledge-building education was no longer our priority. Our future intervention strategies should place greater importance on developing motivation and behavioural skills to achieve consistent condom use among migrant MSM.

Finally, we compared two models (full and restricted) to find which had the best fit for predicting consistent condom use among migrant MSM. The P-value after comparison showed no statistically significant difference in fit between the models when assessing the difference of the chi-square of the fit for each, which was χ²(3) = 3.464, P > 0.05. However, we found that in the restricted model χ²/df and RMSEA decreased, suggesting a better fit. Additionally, the restricted model was more refined and useful for guiding intervention strategies, which suggested we should shift from focusing on information alone to a more integrated strategy collectively including information, motivation and behavioural skills.

Our study had several limitations. First, we collected information through a self-reported questionnaire. Although it was standardized and our results showed good consistency, information bias may still have been present. Additionally, the snowball method was used to identify eligible participants, rather than a random trial. This may limit our ability to generalize our findings to other MSM populations. We should expand the sample size in future studies. Finally, this study focused on internal migrant MSM in Shanghai, which may limit the results’ generalizability to other regions of China.

Despite these limitations, to our knowledge, this research was the first attempt to apply the IMB model to predict condom use behaviour among migrant MSM in China. Our findings indicated future intervention strategies should emphasize the importance of cultivating motivation and behavioural skills among migrant MSM.

The findings of this study indicated low prevalence of consistent condom use among the investigated migrant MSM, and that timely action is needed to prevent further HIV transmission among them. Intervention strategies should be comprehensive enough to include not only transferal of information, but also other dimensions, such as cultivating motivation and behavioural skills.