Surveying Indian gay men for coping skills and HIV testing patterns using the internetKS Jethwani1, SV Mishra2, PS Jethwani3, NS Sawant4
1 Center for Connected Health, Partners Health Care and Massachusetts General Hospital; Department of Dermatology, Harvard Medical School, Boston, Massachusetts, USA
2 Co-founder, Decimal Foundation, Mumbai, Maharashtra, India
3 Department of NGO Relationships, Samhita Social Ventures, Mumbai, Maharashtra, India
4 Department of Psychiatry, King Edward Memorial Hospital, Mumbai, Maharashtra, India
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0022-3859.132315
Source of Support: None, Conflict of Interest: None
Background: Surveying vulnerable and incarcerated populations is often challenging. Newer methods to reach and collect sensitive information in a safe, secure, and valid manner can go a long way in addressing this unmet need. Homosexual men in India live with inadequate social support, marginalization, and lack legal recognition. These make them less reachable by public health agencies, and make them more likely to continue with high-risk behaviors, and contract human immunodeficiency virus (HIV). Aims: To understand coping skills and HIV testing patterns of homosexual men versus heterosexual men. Materials and Methods: An internet based study using a secure web platform and an anonymised questionnaire. The brief COPE Inventory was used to assess coping styles. Results: A total of 124 respondents were studied. Homosexual men used negative coping skills such as behavioral disengagement and tested for HIV significantly more often than heterosexual men. Heterosexual respondents used positive coping skills more often. The most commonly used coping skill by heterosexual men was instrumental coping and by homosexual men was acceptance. Discussion: Overall, homosexual men used negative coping mechanisms, like behavioral disengagement more often. The Indian family structure and social support is probably responsible for heterosexual men's over-reliance on instrumental coping, while resulting in disengagement in homosexuals. Conclusion: The lack of legal and social recognition of homosexuality has negatively impacted lives of gay men in India. This is strongly linked to harmful psychological and public health implications for HIV prevention and mental health for homosexual men.
Keywords: COPE inventory, coping skills, HIV testing behavior, homosexual men, LGBT, online, web-based
Homosexuality has long been viewed as a sexual aberration.  This view exists with varying severity even today in many parts of the world.  Marginalization of homosexual men in India makes it difficult for public health interventions to reach them.  Public health research and interventions require participation and feedback from this community and methods to reach, recruit, and study them rapidly are few.  Homosexual conduct is still a punishable criminal offence as per Section 377 of the Indian Penal Code.  Recent research also shows that India remains deeply homophobic and over 73% of Indians believe that it should continue to remain illegal, while over 83% respondents believe that it is an idea imported from the West.  This deep-seated homophobia makes it difficult for gay men and women to come out, and they are therefore inaccessible to public health researchers and workers. Homosexuals tend to differ from their heterosexual peers, right from their adolescent years, which are wrought with confusion, anxiety, even anger, at not being 'normal'.  All this, in addition to poor social support systems and the increased threat of HIV infection have now increased the incidence of depressive symptoms in members of this community. 
Various methodologies have been employed to collect data from this hard to reach population in the past.  Sampling methods like cluster randomization are ineffective, given the small proportion of the target group.  Survey methods, including face-to-face interviews, informal confidential voting interview (ICVI), random screening, and data collection fail in this group owing to lack of confidentiality and anonymity. Other novel anonymous methods such as programmed color-coded audio computer-assisted self-interview (C-ACASI)  system have demonstrated some success in rural India, but the logistical and financial requirements to setup these programs make them difficult to scale. There is thus a need to explore methods that can be scaled easily and also provide the necessary anonymity, security, and confidentiality of data that is required to access this group. With an exponential increase of internet access across the globe and in India, we explored an internet-based surveying methodology for gay men for coping mechanism and HIV testing and compared it with a group of heterosexual men.
The study was approved by the Institutional Review Board. Anonymity was preserved by using the website www.psychpopuli.com developed specifically for this purpose. The website collected no identifying information; hence was Health Insurance Portability and Accountability Act (HIPAA) compliant. The study is registered with the clinical trials registry of India [REF/2014/03/006568].
The internet platform
Three key tenets were adopted while developing this platform: Data integrity, anonymity, and privacy. On the internet, all visitors to a website leave their internet protocol (IP) address. This address allows website administrators to potentially identify subjects. Identifying and storing this IP address is also the only way to avoid duplicate entries by the same user on the website, since the website collected no other identifying information. However, to maintain subject anonymity from study staff as well as any regulatory authorities, the platform automatically encrypted the IP addresses in an uniform manner making recognition of the encrypted platform difficult. since it encrypted these addresses uniformly, it could still track multiple entries. To maintain privacy, the study was only advertised in fairly anonymous online groups, No information about the study or its purpose was sent to participants on their personal email addresses. Also, if participant logged back in, they would not be able to see the study or data. Participants would not be able to return to the website once they closed their web browser.
The homepage of the website provided brief information about the purpose of the study and details of investigators and their affiliations. The next page contained all requisite information for participants, risks, and foreseeable harm. This was followed by the consent page. Every time a participant read and accepted the provisions of the informed consent form, a time-stamped entry was automatically made in the database. Participants could not proceed beyond this step without consenting to the study. The subsequent pages asked for demographic details (especially nationality and country of current residence), sexual identity, the brief COPE inventory, and details of HIV testing patterns and HIV status. The website was hosted and maintained by the study staff for a period of 6 months, following which the data was downloaded and locked and the website pulled down. Participants were eligible to contact us at any point during or after the study and use the resources page that contained links to related articles and other reading materials on the internet.
Data was collected between 1 st January 2008 and 31 st June 2008.
Participants (Cases and controls)
They were recruited using an Indian online mailing list (Gay Bombay Yahoo group*) that has members of the lesbian, gay, bisexual, and transgender (LGBT) community. A common e-mail was sent to the group, accompanied by a mail by the group moderator, who endorsed the credibility of our study. This email was designed to address all concerns that these members might have about their safety and anonymity and validity of the study, and contained a link to the website. This email was followed by three more reminder e-mails on the group, at one week intervals. The control group of heterosexual men was selected by way of advertisements on a popular social networking site - www.facebook.com that directed willing participants to the website - www.psychpopuli.com as above.
Inclusions and exclusions
Only resident Indian nationals above 18 years of age, who self-identified as male homosexuals (study group) and male heterosexuals (control group) were included in the study. Exclusions were female, bisexual or transsexual individuals, or as persons of Indian origin currently residing in another country.
The COPE Inventory - Assessing coping styles
The Brief COPE inventory,  designed by Carver et al., in 1989 (subsequently modified in 1997) was used for this assessment. It assesses coping along 14 dimensions, with two items per dimension. A brief description of each dimension, along with an example of a question item is presented in [Table 1]. Scores are calculated individually for each dimension. Each item was scored on a four-point Likert scale, with responses ranging from 1 (I don't do this at all) to 4 (I do this all the time).
Post study follow up
After completing the questionnaire, subjects received their scores, with a brief interpretation of their four most commonly used coping skills. They were also provided the option of consulting the investigators for any help or query that they might have in understanding the study or its implication in their life.
Descriptive statistics were used for demographic data. COPE scores were assessed for normality for Kolgomorov Smirnov test followed by parametric or non parametric tests as appropriate. Categorical data for HIV testing patterns were analyzed using the Fisher's exact test. All analyses was done at 5% significance.
A total of 239 respondents answered the survery. Of these, 134 satisfied selection criteria. One hundred and fifteen participants were considered ineligible on the basis of incomplete data, being bisexual/heterosexual women/transsexual, or by virtue of being of non Indian origin [Table 2]. Eighty-nine percent respondents who reached the website, completed the survey. The mean age of the group of homosexual men was 28.9 +/ 7.2) versus 26.48 +/ 8.01) for heterosexual men (P > 0.05).
The analysis of the COPE Inventory suggested that the self-identified homosexual male respondents of our study had coping skills that were significantly different than their heterosexual male respondents. Heterosexual men reported using positive coping skills, such as active coping, instrumental coping, positive coping, and humor significantly more often than homosexual men (P<). Homosexual men, on the other hand, reported using behavioral coping significantly more often [Table 3].
HIV testing patterns
A significantly higher frequency of HIV testing was seen in cases versus controls (P < 0.0001) [Table 4].
This study used an online platform for data collection, with special emphasis on identity protection to overcome the social stigma associated with homosexuality. It is known that poorly adjusted homosexual men are prone to a variety of physical and mental repercussions,  such as indulging more readily in high-risk behavior, like unprotected sexual intercourse, drug abuse,  suicidal tendencies,  etc., as a way of coping with stressors. A thorough understanding of the implications of the Indian culture, traditions, and homophobia on coping skills of Indian homosexual men is lacking and is essential. The methodology used demonstrated promise, since the target population was easily reached and over 89% users who accessed the link to the survey ended up participating in the survey. Furthermore, a significant sample size was reached in a short duration. Gay men and women are increasingly using the internet around the world to find support and communities for themselves, access information about LGBT life anonymously, and to find other LGBT individuals to date. The fear of reactions to such explorations offline, family and cultural pressures, and the preference to maintain anonymity is contributing to increasing this trend. 
A common criticism of using the internet for such surveys is the fear of spurious responses. Although this problem can never be fully addressed, certain technical guardrails can reduce the chances. In this study, we recorded the participants' IP address, and restricted participation to one user per IP address. This way, even if a spurious respondent got access to the website, they would only be able to enter information once. We also restricted advertising the study through only one channel to recruit homosexual men, to ensure that the study website was not bombarded with fake identities or spammers. We also did not offering monetary incentives, making it less attractive to dubious participants. It took participants three distinct steps before they were able to start answering the survey, which ensured that internet 'bots' and spammers could not easily access the surveys.
Literature shows that the lack of acceptance and discrimination over the years has impacted how gay men deal with daily stressors.  Gay men have been known to use negative coping skills, like denial and isolation, as ways to cope with these stressors.  Our results were in line with existing literature and acceptance was the most commonly used coping mechanism by gay men. This further demonstrated that the methodology of collecting data could reproduce results found with more traditional methods of collecting data.
In terms of negative coping skills, gay men in our cohort used behavioral disengagement significantly more than their heterosexual peers. In contrast, heterosexual men used active coping, positive reframing, and humor significantly more than their homosexual peers. They also used instrumental support from their friends and family very often to tide over their stressors.
One of the reasons heterosexual men are more likely to seek instrumental support could be that Indian men are often raised in large joint-families. Some continue to live so even after financial independence. Heterosexual men, thus show greater reliance on social support from their families to cope with their problems as compared to homosexuals who for most occasions are closed about their sexuality and probably cannot share this information with their families. This reduces their ability to rely on family for support.
Behavioral disengagement not noticed in studies done elsewhere, is seen in how some gay men deal with stressors in India. Indian men are often financially independent but still bound to their families. In this situation, gay men may often develop a defeatist attitude towards their sexuality and its associated stressors. This includes for example, marrying a woman and/or not publicly disclosing their sexuality. Using such negative coping mechanisms can cause long-term psychological harm, and pave the way for high-risk behavior. On the other hand, using active coping (responding positively to questions such as "I've been taking action to try to make the situation better", on the COPE Inventory) has shown to slow down disease progression over a 1-year period in HIV-infected homosexual men. 
Understanding why homosexuals use certain coping mechanisms can help therapists ensure adequate coping and adjustment of homosexuals towards their sexuality, before embarking on any other interventions, like HIV counseling or deaddiction program's.  This is because poorly adjusted individuals are more prone to default sooner, like quicker subsequent return to unprotected anal sex as compared to other homosexual men, and decreased drug and medical adherence.  They are also significantly less likely to engage in perceived seroprotective behaviors, such as serosorting, irrespective of their serostatus. 
Our homosexual respondents tested for HIV significantly more often as compared to heterosexual men. The heightened awareness about and perception of being a high-risk group for contracting HIV could be the possible reason for this disparity. Seronegative homosexual men can benefit greatly with therapy that is tailored to these findings and the societal context associated with them. Even after contracting the disease, if HIV positive homosexual men use active coping skills, they have fewer physical and psychological symptoms, and tend to have longer life spans than their poorly adjusted counterparts. 
The most significant limitation of this study was the small sample size of 64 and 62 self-reported homosexual and heterosexual men, respectively. The data was is also 4 years old. However, the basic premise of using the internet for reaching certain vulnerable populations is probably more relevant today. Also, the recent judgment on homosexuality make the data relevant. Our respondents were young urban men and likely more aware about HIV and homosexuality. Responses on potential confounders for HIV testing such as frequency of sexual contact and other high-risk behavior were not collected. The study also did not collect any qualitative feedback that would have allowed a deeper insight into why these men cope differently.
Ganesh Prabhu for developing and managing web platform.
*The Gay Bombay Yahoo group is the official mailing list of the 'Gay Bombay Group'. It has more than 16,000 registered LGBT members and about 400 posts a month. More details are available on: www.gaybombay.org.
[Table 1], [Table 2], [Table 3], [Table 4]