Abstract
The HIV epidemic in Trinidad and Tobago is primarily heterosexual, fueled by a high level of risky sex, gender inequality, and alcohol and drug use; however, the influence of alcohol and drugs has been neglected in the literature. Research shows that current HIV prevention approaches have failed to substantially impact sexual behavior change. This may be so because they do not incorporate a comprehensive understanding of the sociocultural factors underlying sexual behavior. There is an urgent need to understand how socially accepted patterns of alcohol and drug use contribute to sexual behaviors and HIV risk in Trinidad and Tobago. Moreover, specialized, evidence-based interventions are needed for HIV-infected substance abusers. Using an adaptation of the cognitive behavioral stress management (CBSM) protocol, this intervention project aimed to assess effectiveness in reducing relapse and risky behaviors among recovering HIV-infected substance abusers in Trinidad and Tobago.
Introduction
The HIV epidemic is well established in the Caribbean. Most recent estimates (2008) show a prevalence rate of 1.0%, with a range between 0.9% and 1.1% across the region. 1 Approximately 20 000 people were newly infected in 2007. These statistics place the Caribbean second in the world ranking of HIV prevalence and the region most affected in the Americas. An estimated 240 000 persons in the Caribbean are living with HIV/AIDS, almost half of whom are women. Young women are especially vulnerable, having infection rates significantly higher than their male counterparts. 2,3 In 2004, AIDS-related illnesses were the fourth leading cause of death among Caribbean women and the fifth leading cause of death among Caribbean men. 4 In 2006, HIV/AIDS remained one of the leading causes of death in the 15- to 44-year age range. 3
The characteristics of the HIV epidemic in the Caribbean have been well described. 1 –3 In most countries, it is a primarily heterosexual epidemic fueled by high rates of unsafe sexual behaviors and sociocultural norms and religious taboos that hinder safe sex practices; gender inequities in socioeconomic and power which undermine women's potential for self-efficacy; and alcohol and drug use. Poverty, unemployment, poor health care, stigma, sex tourism, and migration have also contributed to the epidemic. Sex between men is a further significant factor primarily because homophobia of varying intensities has rendered men who have sex with men (MSM) an almost invisible subpopulation. Consequently, accessibility to mainstream interventions is minimal.
The epidemic in Trinidad and Tobago is driven by a combination of these factors. With an HIV prevalence rate of 1.5% in 2007, it has one of the highest rates of HIV in the Caribbean. The rate of infection exceeded the 2007 global world prevalence of 0.8% and the rate in North America (0.6%). Tobago is differentially affected, with one report in 2000 estimating a prevalence rate of 3.5% among 15- to 24-year-olds. 5 Reported new cases of HIV infection in Trinidad and Tobago have increased steadily since the early 1980s. 6 The first case of HIV among males was reported in 1983. The first reports among women occurred in 1985 and since then a feminization of the epidemic has been observed. Forty-seven percent of all incident cases occurred in females in 2007, with women accounting for 70% of new cases in the 15- to 24-year age group. 7 The most common route of transmission is through heterosexual sex, but it is estimated that a significant number of new infections are acquired through homosexual sex. As a result, even though HIV transmission is predominantly heterosexual, in Trinidad and Tobago, the epidemic is regarded as being simultaneously heterosexual and homosexual.
High rates of alcohol/drug use and abuse have been reported in Trinidad and Tobago. 8,9 The intersection of substance use and HIV is understudied. In spite of early evidence pointing to an association between crack cocaine use and HIV infection, no in-depth research has taken place, and local funding sources have been unready to support proposals for research in this area. In the early years of the HIV epidemic, descriptions of the drug use/HIV link were mainly influenced by the North American focus on intravenous drug use. However, regular intravenous drug use remains a rarity in Trinidad and Tobago and most commonly occurs among persons who have resided in the United States or other countries where intravenous drug use is more prevalent and who have returned to Trinidad voluntarily or more often, involuntarily. Intravenous drug use is therefore not considered to have a significant impact on the HIV epidemic in Trinidad and Tobago though monitoring should continue to assure early intervention if the need arises. There is also inadequate insight into the relationship between alcohol use and HIV in Trinidad and Tobago and the wider Caribbean. In general, research in the Caribbean is hindered by the absence of an explicit research agenda, accurate data documentation, and appropriate data collection. As a result little empirical data are available on these cofactors of risk.
This review assembles the body of available data on alcohol and drug use, sexual behaviors, and HIV risk, and in so doing highlights knowledge gaps that exist and directions for future research. Data suggest that alcohol and drug use in normal, socially sanctioned usage has the propensity to increase HIV risk. A deeper understanding of the link between alcohol/drug use and HIV/AIDS therefore requires exploration beyond the paradigm of an exchange of sex for illicit drugs. Alcohol and drug use needs to be seen as having a significant influence on sexuality and sexual cultures. It is expected that this review will have significant impact on the further management of the HIV epidemic in Trinidad and Tobago and in the Caribbean, as it highlights the importance of integrating the sociocultural norms of drinking and drug use in HIV prevention planning.
This review also explores the need for effective, empirically validated, specialized interventions for the understudied group of HIV-positive drug abusers. A randomized trial experience of cognitive behavioral stress management (CBSM) will also be examined, in both its promising aspects and its shortcomings in addressing HIV prevention, relapse prevention, stigma, HIV testing, and scale-up challenges in Trinidad and Tobago.
Overview of Trinidad and Tobago
Trinidad and Tobago is a twin island country situated at the southernmost end of the chain of islands in the Caribbean Sea, a mere 11 km (7 miles) from Venezuela at its closest point. Trinidad, the larger of the 2 islands comprises most of the 5128 km2 of the country. Tobago, which comprises approximately 6% of the square area (300 km2), has approximately 4% of the total population of 1.3 million (national census, 2000). Trinidad and Tobago is one of the richest countries in the Caribbean primarily through the production and export of natural gas and petroleum. There is an attempt at diversification of the economy, with an increasing focus on the development of manufacturing, the agricultural sector, and tourism. For Trinidad alone, travel and tourism contributes 10% of the gross domestic product (GDP) and 15% of employment (2009); while, for Tobago, it has a much higher economic significance, generating 37% of estimated GDP and just over 50% of all employment on the island. 10 With a GDP of 23 100 USD in 2009, 11 representing a GDP growth rate of over 100% since 1999, Trinidad and Tobago has one of the highest growth rates and per capita income in Latin America and the Caribbean. It remains financially efficient and is considered a high-income country according to the World Bank classification
This wealth translates to a high standard of living for most citizens of Trinidad and Tobago. The Health Development Index rose from 0.794 to 0.837 between 1980 and 2007. 12 Education is compulsory from age 5 and is free at the primary, secondary, and tertiary levels. Literacy rates are 99% for all adults above 15 years, and 100% for adults 15 to 24 years. 11 The unemployment rate declined from 10.8% to 7.5% between 2003 and 2009. 11 Based on reports from 2000 to 2007, 4% of the population live below international poverty line of US$1.25/d. 12
Trinidad and Tobago is a multiethnic, multireligious society which exemplifies harmony, tolerance, and diversity. The 2 dominant ethnic groups are South Asian Indians (40%) and Africans (37.5%), and 20.5% of the population are of mixed ethnicity. 11 The population is primarily urban (75%). 13 The male to female ratio is 1.02 with 72% of the population between 15 and 64 years, 20% under the age of 15 years, and 8% 65 years or older. 11 Life expectancy is 70.8 years. 11 In 2000, HIV disease was the fifth leading cause of death among all ages and the leading cause of death among the age group 15 to 34 years. 4 In one report (2005), it was projected that in 2010, life expectancy would be 10 years less in Trinidad and Tobago than it would have been without AIDS. 14
Generally, the people of Trinidad and Tobago are regarded as fun loving. 15 Carnival, an annual national festival, is undisputedly the most significant cultural event. Almost 2 months of cultural events highlighting rhythmic music, drinking, and festivities culminate in a 2-day street parade of spectacular costumes, revelry, and euphoria. This 2-day period is almost universally regarded as a time during which the liberation of inhibitions and violation of social norms are acceptable.
Overview of Substance Use in Trinidad and Tobago
Psychoactive substance use is prevalent in Trinidad and Tobago. 16 The most commonly used substances in the adult population are alcohol, nicotine, marijuana, and crack cocaine. 9 Alcohol plays an important social and cultural role in Trinidad and Tobago. Drinking is accepted as a part of the culture even among youth. Alcohol is also a central element of life's rituals including celebratory and transitional rituals. Alcohol serves as a social lubricant, a sign of male bonding, the social stamp of business transactions, and an expression of commiseration. Drinking is expected as part of almost all social activities so much so that its absence may evoke queries of concern. Even in situations where alcohol is formally prohibited, for example certain religious ceremonies, it is not unusual to find alcoholic beverages available at a respectful distance.
In 2003, per capita alcohol consumption for persons 15 years and older was 4.2 L of alcohol. 17 Traditionally, women are expected to consume alcohol in smaller quantities than men do and to consume more "feminine" drinks that are sweeter and weaker in alcoholic content.
Surveys of alcohol consumption and abstinence rates among adults in various districts in Trinidad and Tobago between 1977 and 1995 place abstention rates among men in the range of 29% to 47.1%, while 68.9% to 80.9% of women were found to be abstainers. 17,18 Regular drinking and binge drinking are common among men. Researchers found that 64.6% of men and 24.9% of women interviewed in St James and Woodbrook, urban areas just outside the capital, drank at least 1 drink/week. 19 The median intake of alcohol was 4 drinks/week 20 ; 19% of males regularly drank more than 20 drinks in the previous 7 days. 21 Country reports (1995) from Trinidad and Tobago 22 reported that 80% of adult males and 54% of females had consumed 12 or more alcoholic drinks in their lifetime. This was especially true among those with lower educational attainment. Heavy drinking (at least 21 units of alcohol/week) was reported by 10.5% of males. The prevalence of heavy drinking rose to 13% of men in central Trinidad where the sugar industry is based.
Rates of consumption are equally high among adolescents among whom the most commonly used drugs are alcohol, nicotine, and solvents/inhalants. 16 Apart from peer pressure, the use of alcohol as a medicinal agent and its role in celebratory and transitional rituals are routes through which alcohol use begins in children and adolescents. 23 In one study, 84% of secondary school students aged 14 to 18 years had ever used alcohol, while 42.5% of adolescent males and 11.3% of adolescent females between the ages of 16 and 19 years admitted to drinking at least once per month. 24 A national survey among 13-, 15- and 17-year-olds in secondary schools revealed a lifetime prevalence of alcohol use of 83.7%, with 50% of the sample having drunk alcohol in the month prior to survey. 16 In 2007, the Trinidad and Tobago Global School-based Student Health Survey (GSHS) collected data from 1692 students in secondary schools. 25 This survey also found high rates of substance use and earlier age of onset of alcohol use among students in forms 1 through 4 (corresponding to ages 11-17 years). The prevalence of alcohol use in the month prior to the survey was 40.5%, with no significant difference between the sexes (42.3% males and 38.8% females). Almost half of the male students (48.1%) and more than one third of females (36.4%) reported having 2 or more drinks when they drank; 28.1% had experienced intoxication and 16.8% had adverse effects related to drinking, which affected school, family life, health, friends, or normal social behavior. In spite of legislation prohibiting the sale of alcohol to persons under the age of 18 years, 27.2% of male and 7.3% of female students had purchased alcohol themselves.
Binge drinking has also been reported among adolescents. Boyd Patrick found that acute heavy drinking (more than 5 drinks on a single occasion) was more prevalent among adolescent girls than among adult females. 24 The adoption of newer drinking styles and options among young people has resulted in an increased rate of binge drinking of stronger alcoholic drinks among both males and females. The norms restricting excessive alcohol use among females in the younger generation appear to have been altered with a growing acceptance of female drinking. The physiological disadvantage resulting in the higher blood alcohol levels by female drinkers consuming as much alcohol as men 26 has serious implications for the sexual transactions that are likely to occur. Becoming intoxicated more readily than their male friends, female adolescents are therefore more likely to display impaired judgment, increased risk taking, and have less ability to negotiate safe sex. 27
Binge drinking is a high-risk behavior and predisposes individuals to alcohol abuse and dependence. In 1989, 14.5% of males in Tobago gave responses on the CAGE screening tool that required further evaluation for alcoholism. 28 In one community survey of adults in Trinidad and Tobago (2004), 5% (11.4% male) of those screened with the Alcohol Use Disorders Identification Test (AUDIT) had medium to very high scores (AUDIT score ≥ 8), suggestive of a possible alcohol use disorder. 29 High-risk drinking as an expression of either alcohol abuse/alcoholism or the socially sanctioned norms of drinking also predisposes to risky sexual behaviors.
The most dramatic and explicit demonstration of the excessive use of alcohol and associated sexual abandon occurs at the annual national carnival celebrations. For almost 1 month leading up to the artistic street parades, Trinidadians and Tobagonians are immersed in revelry, public binge drinking, and anecdotally, sexual risk taking, which culminates in the 2-day explosion of street "mas." Alcohol is an intrinsic part of carnival and is a vehicle that facilitates public sensuality and sexuality, exhibitionist displays, varying states of undress, and a removal of the barriers between social classes and sexes. Heavy alcohol use is sanctioned and drunkenness is accepted as an integral part of carnival celebrations.
Literature reviews yielded no official documentation of increased alcohol consumption, increased risky sexual behaviors, or changes in sexual behavior during carnival, but it is unequivocally accepted that this is a time of loosened sexual inhibitions and risk taking, facilitated by alcohol. Carnival is also a time when the country hosts many tourists who are likewise assimilated into the cultural patterns of alcohol use and other behaviors. Even outside of the season of carnival, the projected images of tourism in Trinidad and Tobago and the rest of the Caribbean is one of heavy alcohol use and sex
In contrast with alcohol, marijuana and crack cocaine are classified as dangerous drugs under the Dangerous Drugs Act, which defines possession of any quantity of these substances as a criminal offence. Both these illegal drugs are readily available however. While Trinidad and Tobago is not known for widespread cultivation and export of marijuana, the drug is grown locally and both the local and imported products widely consumed. Tolerance of its use by society is facilitated by the commonly held beliefs that marijuana is harmless and potentially helpful particularly as a medicinal aid. The drug also has gained social acceptance among some because of its association with Rastafarian subculture. In spite of its illegal status, the prevalence of marijuana use among adolescents and adults in Trinidad and Tobago is significant. The 2007 GSHS study of adolescents in high schools 25 reported that 13.6% of students had ever tried drugs including marijuana, hemp, or cocaine. The National Secondary Schools Survey (2006) revealed that 12% of students had tried smoking marijuana (18% males and 7% females), while 0.74% admitted to ever using crack cocaine. 16 A higher percentage of students admitted to using any illicit drug (14.7%), most likely an indication of the popular use of hemp as a more acceptable alternate to marijuana. Hemp is used to describe cannabis plants high in fiber content, whereas marijuana is used to describe cannabis plants high in psychoactive components. Marijuana is also perceived as a euphoriant that enhances sexual pleasure and increases sexual desire. This belief is not specific to Trinidad and Tobago.
While the use of alcohol and marijuana is intricately woven into the cultural fabric of Trinidad and Tobago, the use of crack cocaine is more a consequence of the island state serving as a transhipment point for the drug from the coca-producing countries in South America to markets in North America and Europe. The short distance between South America and Trinidad, the mobility between islands via the sea ways, and the relative absence of adequate coastguard patrol have lent ample opportunity for transfer of cocaine powder. Pure cocaine powder is converted into crack cocaine through a simple chemical procedure that does not require much sophisticated equipment. The resulting easily ignited base of cocaine, when smoked results in an immediate and intense euphoria for a brief 5- to 10-minute duration. A spill off from the use of Trinidad as a transhipment location was the ready availability of the drug to locals as crack cocaine became a part of the local drug market. This led to the escalating prevalence of crack cocaine use and dependence, petty crime by crack cocaine users, serious crime associated with drug dealing, national corruption, and the influence of international drug cartels. Based on anecdotal, clinical, and personal communications, risky sex seems endemic among crack cocaine users, the dynamics of which demand further exploration. Unlike with marijuana, which is perceived to have aphrodisiac properties, available reports indicate that sex associated with crack cocaine use carries overtones of violence, gender disparities, sexual dysfunction, and power. 30
Sexual Behavior, Substance Use, and the HIV Epidemic in Trinidad and Tobago
The first cases of AIDS in Trinidad and Tobago were reported in 1983 among men who have sex with men (MSM). As described by Cleghorn et al, 27 cases of AIDS were diagnosed between 1983 and 1984, all of whom were homosexual or bisexual men. 31 Molecular epidemiologic data and genetic subtyping of the virus have confirmed that the HIV-1 virus was introduced to Trinidad through homosexual/bisexual sexual activity between men from Trinidad and North America. 32 By 1985, AIDS was diagnosed among women and children, signifying the start of a predominantly heterosexual epidemic. 6,31 In 1985, there were 38 cases of AIDS, 5 heterosexual and 33 homosexual/bisexual. All the women were infected by bisexual men. Between 1985 and 1988, the number of cases among homosexuals/bisexuals doubled while there was a 12-fold increase among women. In 1988, there were more heterosexual than homosexual/bisexual cases of AIDS (66 versus 61). The number of HIV cases steadily increased from 8 in 1983 to 626 in 1993 to 1718 in 2003. 6 The northern county of St. George in Trinidad and Tobago were disproportionately affected initially, however all geographical areas were steadily affected as the epidemic progressed. 7 The predominant mode of transmission among adults is through unprotected sexual contact 33 —up to 60% through heterosexual sex and an estimated 15% through homosexual and bisexual behavior. Other modes of transmission are uncommon and include blood transfusions, mother-to-child transmission, and intravenous drug use. Two thirds of all cases are in the 15- to 44-year age group.
Not only was the heterosexual nature of the epidemic clearly established early on, but by 1993 there was a clear trend toward a feminization of HIV infection and AIDS. 6 The percentage of women infected increased significantly from 0% in 1983 to 35% in 2000, and the male to female gender ratio fell from 4:1 in 1985 to 1.4:1 in 2006. Forty-five percent of all new HIV cases occur in females, while 70% of new infections among 15- to 24-year-olds occur in women in Trinidad and Tobago. This latter category is considered the group at highest risk of HIV infection in the nation.
With significant support from international agencies, and through local and regional programmes, the treatment and prevention of HIV/AIDS in Trinidad and Tobago has advanced, so that in recent years the epidemic has been described as stable. Following the implementation of the National HIV/AIDS Strategic Plan (January 2004 to December 2008), the government reports its major achievements to include the availability of HIV testing at all health care facilities, many of which offer same visit results, and since April 2002, free antiretroviral (ARV) medication for all who are in need. The numbers of persons seeking treatment have steadily increased, and the progression from HIV infection to AIDS has steadily diminished. The ratio of AIDS to HIV infection declined from 1:15 in 2005 to 1:8 in 2007. 34 AIDS-related deaths also declined 53% between 2002 and 2006 because of the wider access to ARV medication. 34 HIV prevalence 35 among pregnant women attending public antenatal clinics dropped slightly from 1.9% in 2000 to 1.6% in 2005. Mother-to-child transmission was 3% in 2007 and the prevention program for mother-to-child transmission has resulted in earlier diagnosis of HIV-positive babies. Attention has also been placed on chronic care of persons with HIV-related illnesses, discrimination in the workplace, training and capacity building. Most HIV prevention programs address sexual behaviors through the ABC approach which promotes Abstinence, Being faithful in monogamous relationships, and Condom use. The general population has been the main beneficiary of behavior change messages; programs have been geared to both an adult heterosexual audience and youth of 15 to 24 years old.
A recent review suggests that the general population has demonstrated increasing knowledge about HIV/AIDS, its transmission, and the available resources for treatment. 36 There has also been a reduction in HIV-related stigma and increased testing. The same review however draws the disturbing conclusion that current HIV prevention approaches have failed to substantially impact sexual behavior. Between 1998 and 2007, there was no indication that sexual behaviors had changed. What had changed was an increased awareness of risk and the need for HIV testing. This is not surprising since sexual behaviors are very complex human interactions which require careful dissection and thorough understanding if they are to be changed. Simple directives promoting change in sexual behaviors are therefore inadequate to generate effective interventions. Progress in behavioral prevention strategies will require a more comprehensive understanding of sexuality and the sociocultural factors underlying sexual behavior. In particular, there is an urgent need to understand the role played by the socially accepted patterns of alcohol and drug use in promoting risky sexual behaviors. In general, this is an area that has not received much attention in the evaluation and management of the HIV epidemic in the Caribbean.
The unsafe heterosexual sexual practices fueling the HIV epidemic in Trinidad and Tobago have been described in several studies. However, existing research has basically been geared toward eliciting the prevalence of risky sexual behaviors. Very little has been written explicitly about the motivations and underpinnings of these sexual behaviors in the general population.
Studies conducted among adolescents have shown high rates of sexual activity, early onset of sexual intercourse, multiple partnering, and reluctance to access and use condoms. The 2007 GSHS survey 25 gathered a wealth of data on the sexual behaviors among adolescents aged 11 to 18 years. Twenty-six percent of respondents were sexually active (32% males and 20.2% females) with 13.2% initiating sexual activity before the age of 13 years, and 24.9% having sexual intercourse in the month prior to the survey. Serial partnering was evident with 23.9% of males and 11.4% of females admitting to 2 or more sexual partners in their lifetime. Of the 673 young persons 10 to 19 years old interviewed in Tobago in another study, 60% of males and 36% of females reported sexual activity. 37 Among the sexually active, 20% of the 10 to 14 years old and 25% of those aged 15 to 19 years reported having had more than 5 lifetime partners. The average age of coitarche was 14 years; 6.5% had initiated sex by age 10 years and 25% by the age of 12 years. Males gained prestige among peers by increasing their number of sexual partners.
More than half of the sexually active adolescents (59.1%) in the GSHS survey admitted to using a condom at the last sexual encounter, but only 37.3% said they would seek condoms from a pharmacy or health facility. 25 Findings from focus groups with youth in Tobago indicated the reluctance of youth to use condoms when they felt they were in love and revealed that females deferred to male partners in decisions about condom use. 37 Young females in Tobago constrained by religious norms that demand abstinence were reluctant to buy condoms for fear of social stigma, which would be magnified in a small community. The significant rates of unprotected early sexual activity are borne out by the high rates of teenage pregnancy. The 2000 Population and Vital Statistics Report states that almost 14.8% of live births in Trinidad and Tobago were to teenage mothers.
The disinhibiting effect of alcohol increases socializing but alcohol use also facilitates engagement in what may otherwise be considered undesirable or repugnant behavior. Findings from a focus group study of university students described the use of alcohol by young men who felt pressured to conform to societal expectations of masculinity by engaging in unwanted sexual activity.
38
Alcohol helps a lot of times to give into a woman you don’t really want. Use it (alcohol) on yourself to become bolder in approaching females and to have lyrics.
The women in this study reported the association between the use of alcohol, sexual violence, and nonconsensual sex. Marijuana use is also reportedly associated with risky sexual behaviors. This has not been documented among youth in Trinidad and Tobago; however, among Jamaican adolescents attending a clinic for sexually transmitted infections (STIs), there was a high prevalence of early sexual activity, unprotected sex, and multiple sex partners.
39
Significantly, marijuana use, which was documented in 60% of these adolescents was an independent risk factor for dysuria, a proxy for risky sexual behavior and HIV risk.
Unsafe sexual behaviors are also documented among the adult population where the significance of alcohol and drug use is more frequently reported. Furlonge and colleagues (2000) examined the prevalence of sexual risk behaviors and the precursors to sexual risk taking among a population-based probability sample of 860 persons in North and Central Trinidad. 40 During the 2 months prior to interview, 51% of respondents reported unprotected sexual intercourse with a primary sex partner (ie, spouse or steady partner), and 4% reported unprotected sexual intercourse with a nonprimary partner (ie, casual or commercial partner). Having knowledge of HIV transmission and keeping condoms at home were associated with a lower likelihood of engaging in unprotected sexual intercourse with both primary and nonprimary partners. Alcohol and drug use prior to intercourse was associated with reports of unprotected intercourse with both primary and nonprimary partners. Among persons attending a clinic for STIs in Trinidad, drug use prior to engaging in casual and/or oral sex was significantly associated with gonococcal infection among males. 41 Another smaller survey of 227 persons in the general population also revealed high rates of sexual risk taking. Seventy-four percent of respondents had unprotected sex at least once in the 2 months preceding the survey; 86% had unprotected sex in the 6 months preceding the survey; 9% reported 2 or more partners. Alcohol (38%) and marijuana (12%) use before sexual intercourse was prevalent. 42 The sexual dynamics described seem to persist even after a diagnosis of HIV infection. Following diagnosis of HIV infection among individuals participating in voluntary HIV counseling and testing at a sexually transmitted disease clinic in Trinidad, 33% of men and 27% of women admitted to unprotected sex with their primary partner. Unprotected sex with a nonprimary partner was reported by 18% of men and 25% of women. 43
Undoubtedly, sexual abandon and risky sexual behaviors reach a pinnacle during the carnival season. No statistics could be found, which documented the increase in unprotected sexual activity and the outcome. However, Brebnor in her qualitative work reported the belief of Tobagonians that musical rhythms and lyrics, such as the music played during the carnival season, and the accompanying dances, indirectly influenced sexual behavior by heightening sexual feelings. 44 Even without scientific evidence, it is accepted that alcohol use at this time contributes significantly to unsafe sexual conduct. Anecdotal reports from medical practitioners reported in the national media document the occurrence of increased incidence of unprotected sex and STIs by as much as 25% among young persons between the ages of 16 and 30 years during the carnival period. 45
Epidemiological data have defined specific populations at high risk of HIV infection. Higher rates of risky sexual behaviors have been reported among these subpopulations in Trinidad and Tobago. Crack cocaine users have been identified as one such high-risk group. 30 Use of crack cocaine has been associated with sexual behaviors that increase susceptibility to HIV infection, not just among users, but also among their personal and commercial sexual partners. 30 There are reports of high rates of multiple partners and unprotected sex. Sex-for-drug and sex-for-money transactions are also a part of the subculture of crack cocaine users. This exchange of sexual services for drugs or money to purchase drugs has been described, 46,47 but it has not been adequately explored and scientifically documented in Trinidad and Tobago. Women are particularly affected, since they are more likely to engage in prostitution to sustain crack cocaine use. 47,48 Among the homeless, women were more likely to be HIV positive or perceive themselves at high HIV risk and to be sex traders. 48
There have been fewer descriptions of patterns of sexual behaviors among other high-risk groups and no real exploration of the influence and role of alcohol and drugs on sexual behaviors among these groups. Men who have sex with men are regarded as a high-risk group. HIV rates among MSM are reportedly between 20% and 40%, much higher than in the general population. 49,50 De Groulard cautions against the impact of homophobia on reported HIV prevalence rates among self-declared gay men who are a minority among MSM and suggests that any prevalence rates are likely to be an underestimate. 51 Male homosexual behaviors are not tolerated in Trinidad and Tobago society; homophobia and stigmatization are potent. The potency of the stigma is intensified by the initial perception of AIDS as a homosexual disease, as well as the fact that in Trinidad and Tobago, as in most Caribbean countries, sodomy is a criminal offence. Female homosexuality is better tolerated and is frequently more regarded as behavior for the titillation of men's sexual excitement, rather than a lifestyle. For example, in focus groups among university students, males approved bisexual behavior among females once the male is also included. 38 The attitude to male homosexuality was much less tolerant. Men therefore do not readily identify with or admit to a homosexual lifestyle or homosexual behaviors. In one study, among MSM, 25% said they regularly also had sex with women. 50 The implications of this bisexual behavior have been documented throughout the HIV epidemic in Trinidad and Tobago. The use of alcohol in this context has been described by Cáceres 52 in his commentary on MSM in Latin America and the Caribbean, "… the intense use of alcohol (heavily promoted in leisure contexts in the region as a facilitator of social interaction, and even signifying ‘normal’ male behaviour) is as much an excuse for the expression of feelings, particularly sadness, weakness and affection for other men, as it is a relatively well-accepted explanation for homosexual interactions across all classes.".
Men who do not identify with a homosexual lifestyle, however, do admit to engaging in male homosexual behaviors (MSM) under particular circumstances. This occurs most commonly among male crack cocaine addicts who trade homosexual sex to acquire drugs or money for drugs, 30 those under the influence of alcohol 53 and men involved in commercial sex work including sex tourism. 54
Tourism is an important part of the economy, particularly in Tobago. The heavy reliance on tourism among Caribbean countries has given rise to these destinations being havens for sex tourism. Sex tourism has increased demand for commercial sex and has been cited as a contributor to the high prevalence of HIV infection in Tobago. Sullivan (2005) in a Congressional Research Service report for Congress commenting on increasing rates of HIV infection noted that "… the Caribbean is a popular destination for sex tourists, and several countries, such as the Dominican Republic, Jamaica and Trinidad and Tobago, have reported that sex tourism is linked to rising infection rates in certain areas." 52 For some, alcohol and drug use is a part of the attraction to the Caribbean as a tourist destination. Tourists who do not have the overt intention of engaging in tourist-related sexual behaviors may use in excess of their usual intake but still remain consistent with local norms. These tourists, under the influence of alcohol and drugs, may be at high risk of engaging in unintended and risky sexual encounters as reported in other regions. 55 One researcher has observed in Tobago that the freedoms associated with being on vacation on an “island paradise,” in conjunction with the liberal use of alcohol and drugs, and the subsequent impaired judgment, perpetuates risky sexual behavior including casual sex with multiple partners and the use of commercial sex workers (CSWs). 44
Although CSWs have been identified as a risky group for acquisition and transmission of HIV, their alcohol and drug use patterns, especially as it relates to their trade have not yet been fully investigated. It is important to determine the level of risky sexual behavior among women who engage in sex trading and the extent to which it is driven by the need for drugs to satisfy addictive urges. Further, there has been little investigation in the HIV prevention literature of the interrelationships between tourism, festive national celebrations like carnival, sex/romance tourism, and commercial sex with alcohol/drug use, and HIV risk. However, research findings in other Caribbean territories may give an indication of the situation.
Among sex workers in the US Virgin Islands, almost three quarters admitted to a history of alcohol or drug use in the month prior to interview, 25.7% reported no substance use, 57.4% reported alcohol use only, and 16.8% indicated use of an illicit drug. 56 Drug-using sex workers reported a significantly greater number of past-month sexual partners than alcohol-only and nondrug users and higher rates of unprotected sexual activity. These risky sexual behaviors were reported in conjunction with STIs, violent victimization, and migration between high and low HIV prevalence areas. In Guyana, 31% of female sex workers were found to be HIV positive. 57 Those workers who regularly got high on alcohol were more likely to be HIV positive (51.2% vs 28.2%) and women who used crack cocaine were also more likely to be HIV infected (51.2% vs 28.2%). The study concluded that HIV rates are particularly high among CSW who use crack cocaine.
The effects of psychoactive drug use in altering judgment and inhibition may lead to impulsive and unsafe sexual behaviors. The studies conducted in Trinidad and Tobago on sexual behavior provide data which suggest that the prevalence of risky sexual behavior is higher when alcohol and drugs are used even when there may not be a substance abuse problem. Among both adults and the youth, in whom unsafe sexual practices seem highly prevalent, the use of alcohol/drugs may erode the impact of any prevention intervention that merely promotes condom use, by altering judgment and the decision to use condoms. This negative effect would be aggravated if the rates of alcohol and drug use are high in a society, as they are in Trinidad and Tobago.
Given the patterns of sexual behaviors among those groups of persons described as high risk, very little has been done on the national level to target these populations of MSM, CSWs, or crack cocaine addicts. Moreover, the patterns of alcohol and drug use among MSM and CSWs need better surveillance. The absence of such information to inform interventions remains a major deficit in HIV prevention in the country.
Crack Cocaine Use and HIV Infection
The association between crack cocaine use and HIV infection has been reported in several small studies in Trinidad and Tobago. 58 –62 From as early as 1988, it was reported that substance abusers seeking treatment at the national Substance Abuse Treatment and Prevention Center were at a higher risk of HIV infection than the general population. Lewis and Hospedales assessed HIV seroprevalence among treatment seekers with crack cocaine dependence. 58 Approximately 3.3% of crack cocaine-dependent men tested HIV positive and 14.3% of the women, a rate 8 times as high as the national rate at that time. Among psychiatric inpatients, being HIV positive was associated with substance abuse, most particularly crack cocaine abuse. 59 The Caribbean Epidemiological Center reported in 1999 that 30% of newly infected persons reported cocaine use. 60 Evidence continued that female crack cocaine addicts were a bridge population for HIV infection in the society. In a review of substance abusers admitted to an all-female rehabilitation center between 1996 and 2002, the HIV prevalence rate was 19.8%, which was 6 times higher than in the general population. 61 HIV-seropositive status was significantly associated with the use of crack cocaine.
Pharmacologically, in the early stages of use, crack cocaine is known to increase sexual desire and heighten sexual arousal while impairing judgment. The preferred method of use in Trinidad and Tobago is freebasing of crack cocaine rocks, which carries a very high risk of rapid progression to addiction. Use of large quantities and prolonged use are more commonly associated with sexual difficulties ranging from impaired desire and reduced pleasure to impotence and delayed ejaculation in men. Chronic use or dependence on crack cocaine is also associated with higher levels of sexual activity, which have less to do with the pharmacology of the drug and more to do with the high-risk setting of crack cocaine houses and dens, and the acquisition of the drug.
The exact mechanism underlying the association between crack cocaine use and HIV infection is not assumed to be a simple one but is mediated in part by the risky sexual behaviors associated with the use of crack cocaine. Compared with their noninfected counterparts, HIV-positive substance abusers are more likely to have had in excess of 10 sexual partners over the past year, with around one third of female crack cocaine addicts admitting to prostitution. 58 Djumalieva and colleagues in their qualitative work sought to explore some of the dynamics underlying these increased risky sexual behaviors, based on interviews with drug users seeking rehabilitation. 30 Risky sexual behaviors were common, with the exchange of sex for money or drugs being described as a normal practice in the crack houses and blocks where crack cocaine was used. Users admitted to "careless and promiscuous" sex during active use, which included multiple partners, same sex exchanges, and unprotected sex. Male crack cocaine users had a preference for oral sex considering it a sexual practice associated with reduced HIV risk. Other studies have reported sexual impotence as a reason for the preference of oral sex, 63,64 but this was not reflected in the Trinidadian sample. One is left to speculate whether this represents a true difference in causation or is a masked reflection of the sociocultural expectations of male virility. Risky sexual behavior was reportedly driven by the need to acquire the drug rather than sexual desire and female crack cocaine users were perceived as being at very high risk of HIV infection because they maintained their habit by exchanging sex for drugs and "would do anything and have sex with anyone in order to obtain drugs." This however did not always deter engagement in sexual behaviors, especially oral sex, with female users. Fatalistic attitudes were also expressed among crack cocaine-using women who had become HIV infected. It was believed that these women gave up on life, rejected rehabilitation, and persisted in unsafe sexual behaviors because of a perception of inevitable death. Risky sexual behaviors were not limited to the population of crack cocaine users, since nondrug-using men accessed services of drug-using prostitutes. This highlights the potential for crack cocaine addicts to significantly impact the progression of the HIV epidemic.
The safety of oral sex with regard to HIV transmission, as perceived by Trinidadian crack cocaine addicts, has not been borne out in the scientific literature. Unprotected oral sex has been associated with HIV transmission especially when oral ulcers, often caused by the use of crack cocaine pipes, or oral or genital ulcers, caused by STIs, are present. 65,66 Having a history of an STI was significantly associated with a positive HIV serostatus among the homeless, 48 psychiatric inpatients, 59 substance abuse treatment-seeking women, 61 and clients of STI clinics. 62
In a case–control study, Cleghorn and colleagues examined HIV seroprevalence among heterosexuals attending an STI clinic and reported an association between HIV infection and crack cocaine use for both males and females. 62 Sex trading, multiple partners, and early coitarche were significant risk factors for HIV infection among women. The women, whose median age was 23 years, were more likely to be HIV positive if they had ever received money for sex (odds ratio [OR] = 5.4, 95% confidence interval [CI] = 1.1-29.3), had more than 5 sex partners in the past 5 years (OR = 5.2, 95% CI = 1.3-21.4), were of younger age at first sexual contact (OR = 4.6, 95% CI = 1.2-21.8), or had nongonoccocal cervicitis (OR = 4.1, 95% CI = 1.3-13.1).The highest odds ratio for males was related to crack cocaine use in the previous 6 months (OR = 6.2, 95% CI = 2.7-14.2) followed by current genital ulcer disease (OR = 5.2, 95% CI = 2.2-12.5), current genital warts (OR = 3.9, 95% CI = 1.2-12.0), and a history of syphilis (OR = 3.2, 95% CI = 1.6-6.1). Similarly, among attendees of STI clinics in another study in Trinidad and Tobago, casual sex was the only predictor of gonorrhoeal infection. 41 Infected males were more likely to have concurrent partners, to have at least 1 sexual partner in the prior 6 months, and to have engaged in drug use prior to engaging in casual and/or oral sex.
In their 2-year review of admissions to the sole psychiatric hospital in Trinidad and Tobago, 6.8% of patients tested were found by Hutchinson and Simeon to be HIV positive. 59 Of these, 58.5% had diagnoses related to crack cocaine use, supporting the significant association between crack cocaine use and HIV infection and introducing the further increased risk of psychiatric illness.
Research has also found a relationship between unsafe sex and alcohol and drug use among the homeless. 46 Among out of treatment, homeless drug users in St. Lucia and Trinidad, 43% admitted to the exchange of crack cocaine or money for sex with high rates of unprotected sex during these transactions (60%). Ninety-five percent of the sample reported crack cocaine use in the past month; self-reported HIV seroprevalence was 25%.
Although research on crack cocaine use in Trinidad and Tobago is not extensive, it is nonetheless sufficient to illustrate how this drug initiates a cycle of increased risk beginning with increased sexual activity, which in combination with impaired judgment leads to sexual risk taking. With a high propensity for addiction, users of crack cocaine, especially female users, readily become involved in higher levels of sexual risk taking, including sex trading, as a way to support an addictive drug habit. These behaviors result in a further direct increase in risk of HIV and other STIs. HIV risk is amplified when there is a comorbid sexually transmitted disease, especially involving ulcers.
The vulnerability of women is further exacerbated by other social factors, for example poor educational accomplishment and early sexual abuse. A deeper understanding of the complex dynamics of sexual activity related to crack cocaine use is necessary, and appropriate interventions implemented. HIV prevention interventions for crack cocaine users need to go beyond condom use and appropriate management of STIs and reflect the apparent intersection between crack cocaine, gender roles, reproductive health, poverty, and HIV.
Approach to the Problem
Sexual behaviors in Trinidad and Tobago are influenced by a sociocultural background of excessive alcohol use and, to a lesser extent, use of marijuana. Over the last decade, data have shown that, especially among adolescents, rates of alcohol use have been increasing, with earlier initiation of substance and sexual activity. Along with these trends, adolescent females are at ever increasing risk. Similarly, over the last decade, rates of HIV have been increasing among adolescents, particularly adolescent females. HIV prevention interventions must be built on a comprehensive understanding of the role played by alcohol and drug use and the compromising effects of psychoactive substance use on sexual behaviors. HIV prevention when focusing on alcohol and drug use must also take into account the sociocultural norms of drinking and consider the practicality of its goal—advocating abstinence in a culture that so completely endorses alcohol use is unlikely to be effective. Among nonproblem drinkers and recreational marijuana users, the focus needs to be one of minimizing harm through a knowledge of the association between alcohol and drug use and risky sexual behaviors.
Reducing HIV risk among the more vulnerable population of crack cocaine users in Trinidad and Tobago has not received priority commensurate with the increased risk noted among this population. Among this population there is no behavioral surveillance, no evidence of change in sexual behaviors, or that there is an awareness of the relationship between crack cocaine use and the risk of HIV infection. This group of persons therefore continues to be at high risk of contracting and transmitting HIV to noncrack users. Reducing risky sexual behaviors will reduce the threat of spread of the HIV virus. Such a harm-reduction approach to HIV risk reduction among crack cocaine addicts is not only feasible but more practical and more likely to have positive effects than the current approach of treating the addiction. The theoretical premise of the latter approach seems to be that if a person is no longer using drugs, risky sexual behaviors will decline and presumably the person will now be amenable to the HIV prevention campaigns that target the general population. There are many flaws in this hypothesis. Abstinence does not necessarily lead to reduction in sexual risk taking among crack cocaine addicts; the rate of relapse among crack cocaine addicts is high; only small numbers of crack cocaine addicts seek and complete treatment; and the popular approach of most HIV prevention programs—abstinence, monogamy and condom use—has not been shown to have any significant impact nationally; it is therefore not expected to be any more effective among the population of crack cocaine addicts in recovery.
Crack cocaine addicts generally do not seek health interventions. A survey of out-of-treatment crack cocaine addicts in Trinidad and St. Lucia revealed significant rates of health care needs especially among homeless crack cocaine addicts, only 20% of whom reported sufficient access to general medical services. 46 Among female substance abusers seeking rehabilitation, high rates of health care needs were also identified. 61 Similarly, fewer crack cocaine addicts seek drug rehabilitation than those who do not, leaving the majority with no targeted HIV prevention if the only intervention is through rehabilitation. Harm-reduction policies and interventions will focus on reducing sexual risk behaviors even among actively using addicts. Such interventions have been useful in other places, such as North America where drug abuse treatment is a useful HIV prevention strategy established for opiate abusers on methadone maintenance treatment. 67 Methadone maintenance reduces HIV risk behaviors through harm reduction focused on the use of clean needles and other paraphernalia for intravenous drug use. Drug rehabilitation programs in Trinidad and Tobago emphasize abstinence as the primary goal and include long-term follow-up usually through community-based 12-step self-help programs. The efficacy of abstinence-based drug rehabilitation programs has not been as clearly established as an effective intervention to reduce HIV transmission, so this should not be the primary approach for HIV prevention among drug addicts.
Even when rehabilitation attempts are made, treatment centers have no clear policy with respect to HIV testing, HIV risk reduction strategies, or the content and format of delivery of HIV prevention. Given the issues surrounding HIV risk among crack cocaine addicts, when addicts seek rehabilitation, relapse prevention is very important, however the critical problem of HIV transmission and safe sex behaviors also has to be addressed. The greater challenge of attracting and retaining women in rehabilitation must also be a focus of attention. HIV-positive women seeking rehabilitation were more likely to be poorly educated, unemployed, and mentally ill, 61 all deterrents to accessing substance abuse treatment.
The point remains that for those in whom relapse proves inevitable, as well as for treatment noncompleters and those never accessing treatment, approaches that focus on the initiation and maintenance of safer sex practices are vital. This public health approach which emphasizes harm reduction will target out-of-treatment addicts and aim specifically at promoting safe sex among active drug users. It should of necessity take into consideration some of the other factors, apart from the need for the drug, that drive risky sexual behaviors among crack cocaine addicts. These factors need to be determined.
In addition, interventions should cater not only to HIV prevention but should also be directed toward crack cocaine users already infected with HIV. Cocaine has been found to be immunotoxic with the potential to significantly decrease CD4 count production by as much as 3- to 4-fold and increase the rates of HIV reproduction up to 20-fold. 68 Alcohol reportedly alters the susceptibility to HIV by enhancing transmission of the virus. 69 When found to be HIV positive, continued crack cocaine use compromises the medical management of the user. Actively using crack cocaine addicts are also more prone to noncompliance with the rigid scheduling of ARV medication which may lead to the emergence of resistant HIV strains with potentially deleterious implications for the management of the epidemic nationally. No data are available as to whether HIV-infected crack cocaine users in Trinidad and Tobago have higher rates of nonadherence to ARV medication, drug-resistant strains of HIV, more severe HIV illness or progression, or medical problems related to interactions between antiretroviral and alcohol or drug use. These are all important areas relevant to health promotion and continued drug use among HIV-positive crack cocaine users who are on ARV medication.
Such harm-reduction policies and strategies will prove beneficial to both the general population and the people using substances themselves. There is therefore a growing need for the development and implementation of effective specialized HIV prevention and treatment interventions specifically targeting the population of crack cocaine users in and out of treatment.
Cognitive Behavioral Stress Management Experience
As previously stated, with the growing recognition of crack cocaine drug users as a high-risk group for HIV infection and transmission, there has been no parallel provision of HIV prevention interventions specifically targeting this population. Interventions have to date been limited to the provision of drug-abuse treatment. In exploring the association between substance abuse treatment and sexual and drug use risk behaviors among HIV-positive persons, one study (2005) concluded that "although the opportunity exists to address HIV risk behaviors in the setting of substance abuse treatment, effective institutionalization of this challenging behavior change effort has not yet been realized." 70 This statement precisely summarizes the situation that exists in Trinidad and Tobago, where there is a clear need for specialized, evidence-based, effective HIV prevention interventions among substance abusers. It was with this need in mind that researchers at the University of the West Indies in 2005 accepted the opportunity for the pilot implementation and evaluation of a risk-reduction program for the population of HIV-positive substance abusers in recovery.
Interventions using CBSM to reduce HIV risk have been proven effective among several populations of persons including alcohol and drug abusers. When adapted and applied to alcohol and drug users, this intervention addresses issues of relapse prevention, stress management, safe sex practices, HIV disclosure, and ARV adherence. Cognitive behavioural stress management effectively reduces HIV transmission and risk of relapse to drug use among HIV-positive drug abusers in recovery. The evolution and effectiveness of this empirically validated, manualized treatment method has been described in detail elsewhere. 71 The pilot project in Trinidad was supported by a supplemental grant to the National Institute on Drug Abuse (NIDA) parent grant # R01 DA13802 “Cognitive Behavioral Treatment for HIV+ Recovering Drug Abusers,” which was executed through Florida International University.
The pilot project had 2major aims. First, the adaptation of the manualized intervention. This included making it culturally appropriate by taking into consideration not just appropriate language and context, but also the other sociocultural factors that underlie alcohol and drug use, and sexual behavior in the Trinidad culture. The adapted intervention would be validated by relevant stakeholders, using focus groups. Second, the project would then aim to apply the adapted intervention in the local setting and assess its effectiveness in reducing relapse and risky sexual behaviors among HIV-positive alcohol and drug abusers in recovery. This project therefore had the potential to make a significant impact by validating a specialized treatment of a high-risk dually diagnosed population, promoting its implementation, and also by increasing the research capacity and exposure among the persons involved in its execution. The project was particularly welcomed since it had been previously established as an effective research-based intervention manualized and therefore easily replicated. It was to be validated for use in the local culture using a randomized controlled trial, and its evaluation would have represented a move toward the implementation of an evidence-based intervention.
From its inception, it was appreciated that a randomized clinical trial among a population of hard to reach, underserved persons would require commitment from the senior clinical staff in all of the potential recruitment sites, as well as other stakeholders. A great deal of effort was therefore expended in gaining support for and commitment to the project. Support from the various agencies was forthcoming once it was clear that the targeted persons were not going to be taken advantage of but that they would be clear beneficiaries. Persons working with the population generally agreed that the intervention was valuable and highly needed. The prospect of a clearly described delivery system with opportunity for replication was well received.
The support was reflected in the attendance and participation in the stakeholders focus groups. The focus groups served 2 main purposes—to apprise stakeholders of the project and further gain their support, and to get from these key informants data that would facilitate the conduct of the research including the adaptation of the assessment and intervention materials, and obtain validation of adaptations that had already been implemented. Agencies represented included HIV treatment services, substance abuse treatment services, HIV support groups, including those for MSM, and drug treatment support groups. Representatives of these agencies were apprised of the details of the project and feedback sought as previously described. Two groups were conducted with 14 attendees from 9 agencies, only one of which had had prior involvement in HIV research. In Trinidad and Tobago, research had not been perceived as a high priority and it was not highly valued among service providers. A randomized trial of a behavioral intervention represented a new level of HIV research in Trinidad and Tobago beyond seroprevalence and descriptive epidemiological studies. The focus group participants and other stakeholders who became involved in the project were ready to embrace this uncharted area.
Even in the absence of empirical data on the challenges of managing the target population, the need for an appropriate intervention was acknowledged, and support for the project was therefore keen. Anecdotal reports abounded from varying stakeholders as to the challenges posed in managing this group of dually diagnosed individuals. Instructive information was obtained from the participants and this was used to fine-tune the project details. From the perspective of substance abuse treatment, themes raised included noncompletion of treatment programs, the revolving door phenomenon of repeated treatment-seeking after relapse, and the poor commitment to long-term follow-up. The success rate after treatment, though not quantified, was perceived as very low. Sustained abstinence among HIV-positive substance abusers was considered to be even lower. One participant even suggested that the majority of HIV-positive substance abusers were homeless and would therefore not be easily reached by the project. The challenges anticipated mainly related to difficulties in recruiting participants who met the project research criteria of at least 30 days abstinent. It was also felt that participants might not be able to sustain abstinence for the 10-week intervention which may have an impact on compliance with intervention sessions.
There was extensive discussion on strategies for recruitment that were likely to be successful and appropriate inducements to sustain interest in participation. To optimize recruitment efforts, stakeholders agreed that the characteristics of the recruiter were important, the approach should be 2-pronged and the inducements should make participants feel that there was something in the whole process from which they could benefit. It was recommended that the recruiter should be someone with knowledge of and exposure to the population and the areas they occupy, someone who could engender trust and make a nonthreatening connection with potential participants, and someone regarded as being able to maintain confidentiality. It was further suggested that prospective participants should be recruited both from the various agencies that provide services and through fieldwork, with HIV-positive substance abusers in recovery spreading the word with a snowballing effect. With regard to inducements, it was unanimously felt that monetary rewards were most appropriate. Many felt that any other forms of incentive would be easily sold if participants wanted money. The distribution of the financial inducement would be structured so that attendance to each subsequent session would bring a greater reward, and a bonus would exist for completion of all sessions.
Concerns from the perspective of HIV treatment providers centered on the unreliability of substance abusers in attendance and compliance with HIV medical management. Fears were expressed of the development of resistance to medication, as the group of substance abusers who were receiving ARV medications were notorious for noncompliance. No group of stakeholders reported that any specific intervention had been introduced into their facilities to treat these problems. There were no management policies in place which addressed them and no liaison between substance abuse treatment facilities and HIV treatment facilities outside of referral from one facility to the other. The occurrence of routine HIV testing depended on the leadership of the specific substance abuse treatment center. A few facilities had gone the route of routine testing but pretest counseling had not been formalized and follow-up did not include any component to specifically address sexual risk behavior. The idea of an intervention that would address all of these HIV-related issues faced by this population was well received by the stakeholders represented. The fact that it was being launched by the local academic institution in collaboration with and therefore supported by a US university, which brought with it international funding and the implications of good scientific practice and high research standards, did little to counter the skepticism that any intervention, research based or not, would make a difference. Pessimism was expressed over the low volume of potential participants and the instability of the target population. Support was pledged nonetheless and additional potential recruitment sites were offered. Several characteristics of potential participants were considered as likely to prompt refusal to participate—fear of stigma, denial of HIV diagnosis, withdrawal from society to maintain recovery, and high risk of relapse.
The cultural adaptation of the treatment intervention and assessment instruments was completed by local researchers to the satisfaction of the researchers in Trinidad and Florida, with no major changes in concepts. Drug use, relationship patterns, and culturally acceptable terminology were incorporated into the intervention; sociocultural norms, values, beliefs, and myths were applied to role-play scenarios and exercises; and relevant health promotion videos were obtained for the control intervention. The adapted intervention was validated by the focus groups and further adapted to include comments derived from discussion and feedback with the 2 groups that comprised academic, clinical, and community representatives, including HIV-positive persons and substance abusers in recovery. There were no HIV-positive persons in recovery in either of the focus group.
Assessors and interventionists for the project were readily identified. Advantage was taken of prior training and experience in the addiction and HIV fields but selections were also guided by competence, passion, flexibility, and locality. Passion would have been necessary to sustain enthusiasm and commitment to the project which represented an additional work demand with minimal reward. Representation from diverse localities was necessary in order to have a broader catchment area for the recruitment and assessment of participants. Assessors especially needed to be flexible and this was accommodated by their employers who had given support to the project. The project coordinator was also trained and proved to be a steady source of motivation and expertise. It was interesting that higher levels of interest, commitment to research, and clinical involvement with HIV-positive persons or substance abusers as part of their normal responsibility were significantly influential in determining delivery performance in the research project.
It was resolved that the reproduction of the intervention would be faithful to the original. Adequate investment was therefore made in preparation and training consistent with the mandates of the parent project. The parent project team was hosted to conduct a 2-day training seminar in Trinidad. The local project team was enthusiastic and responded well despite the intensity of the training. The prior investment in gaining support from senior stakeholders facilitated their ready agreement to release staff members who were project team members from their regular responsibilities in order to attend training sessions. Four interventionists, 6 assessors, and the project coordinator were engaged in an overview of the project and trained to conduct the many assessments and to conduct the CBSM intervention as well as the control intervention.
A significant amount of time and effort was expended in the continued training which required quality control through assessment of tape-recorded sessions. Quality control was provided at both the local and parent institutions and an additional assessment training was conducted by the parent project team 4 months after the original. Rigid standards appeared inflexible as reviewers of the parent project deemed culturally influenced exchanges during the assessment process to be inappropriate. This contributed to frustration on the part of the assessors who admitted to experiencing conflict between the caring, nurturing approach of a care provider, and the importance of maintaining objectivity to reduce bias and ensure adherence to a prescribed script. Ongoing critiques contributed to declining morale as assessors battled to maintain the research standards in nonresearch-oriented service settings. From the perspective of the assessors, they were deviating from culturally normative behaviors and complying with the alien culture of scientific rigor. For example, nonverbal yes–no responses from a participant would evoke a verbal reinforcement by the assessor for the purpose of documentation on the tape recording. Feedback that would criticize the leading effect of the verbal response was regarded as unjustified and hypercritical. The constant practice and recording proved quite demanding on not just time but enthusiasm and performance. Training tapes declined in number and the parent institution became very concerned that persons doing the most assessments were not submitting training tapes.
Training of recruiters occurred simultaneously in the first 2-day training workshop. Five potential recruiters were coached by the recruitment officer of the parent project team and a recruitment plan formulated which included feedback from the focus group. As an outcome, 2 recruiters were officially employed by the project. The selection was based on interest, competence demonstrated during the training workshop, intimate knowledge of, or experience with, the study population, access to and familiarity with areas frequented by the target group, and flexibility. Two main recruiters, 1 male and 1 female were originally employed. During the focus group meetings, a potential recruitment coordinator was discussed—someone with extensive experience working with HIV-positive persons, who knew the field well, had ready access among substance abusers, and was known and respected as a treatment provider in the field. It was hoped that this person would have become involved in the project as recruitment coordinator but this individual was unable to commit to the time needed for the project.
The location for assessments and intervention/control group sessions was determined taking into consideration accessibility, physical comfort, and centrality. The spaces used were in a nonthreatening, nonstigmatizing environment, conducive to participation in group activities, with ready access to emergency medical services. Assessors offered flexible scheduling and the day and timing of interventions were generally determined by each individual participant. Interventionists were paired to form teams of both genders, a trained professional, and a paraprofessional with experiential knowledge and clinical experience. Based on the subjective response of participants, intervention sessions were useful and enjoyable. Participants for the most part attended sessions regularly and actively participated in sessions. They understood the evaluative component of the project and were very cooperative. The general atmosphere of sessions was positive and upbeat.
The most challenging aspect of the project was the recruitment of participants. Numerous problems were encountered in obtaining the numbers of HIV-positive substance abusers in recovery to meet the required sample size requirements for the project. This was in spite of recruitment plans in anticipation of such a problem. Major recruitment sites were substance abuse treatment centers and follow-up support groups, HIV treatment providers, and HIV support groups. Each location posed unique challenges that reflected the challenges associated with managing both the substance abuse and HIV epidemics in Trinidad and Tobago, and that were captured in discussions in focus groups with the stakeholders.
The magnitude of the problem associated with recruitment of HIV-positive substance abusers in recovery was never underestimated. The potential challenges were discussed extensively among project team members, stakeholders, and other researchers who had worked with the population. Recruitment and retention strategies were designed anticipating the challenge and taking into account the feedback from the key informants. Particular attention was paid to selection and training of recruiters, establishment of a clear recruitment strategy which would outreach to both the population that was already in the HIV and/or drug rehabilitation health care services and those who were out of treatment. Adequate incentives were put in place both for the recruiters and for the potential participants. Incentives were financial as well as through frequent reminders and maintenance of tracking sheets for close monitoring. Nonetheless, numerous challenges were encountered in the rate of recruitment and great difficulty was experienced in recruiting potential participants quickly enough to meet the projected timeline of the project.
The initial recruitment plan was 2-pronged. Based on feedback from stakeholders, it was decided that the main strategy would be the recruitment of persons who met research criteria and were already registered in either a substance abuse treatment program or an HIV treatment facility. These would be identified by the staff members of the various centers who, through investment in the project, would facilitate participation. At the time of the stakeholder focus groups, 33 potential participants were immediately brought to mind by the staff of the facilities represented. The goal of recruiting the 64 persons needed for the project therefore seemed feasible. The recruitment plan assigned recruiters to key persons at each of the facilities represented as well as to others that were identified during focus group discussions. The other approach to recruitment was designed to access spaces where members of the target population were likely to be found. This approach allowed especially for the inclusion of traditionally excluded subgroups of the target population, such as MSM. Field recruitment was more likely to access HIV-positive persons who were still using drugs. The recruitment plan therefore included referral to rehabilitation. Three drug treatment providers committed to provide potential study participants with immediate access to rehabilitation, encourage participation in the project, and keep these persons in the drug treatment program until the time of admission to a research intervention group.
Even with these plans, potential participants still proved difficult to access and recruit. At the substance abuse treatment centers, the HIV status of alcoholics and addicts was frequently not known. When HIV testing was done, being informed of an HIV-positive status was associated with a high risk of absconding from the program and a return to drug use. For the small pool of persons meeting project criteria, recruiters experienced difficulty meeting with clinical staff at some treatment centers and were therefore unable to benefit from the personal knowledge that these staff members would have had about potential participants. Recruitment from waiting rooms using flyers and pull-off tabs evoked interest among those seeking care but did not result in successful recruitment. At the national substance abuse treatment center, one recruiter made direct contact with a large group of substance abusers in recovery giving an overview of the project, its relevance, and importance. Recipients were interested, acknowledged the usefulness of the information, and the benefits accrued as a result of the discourse, but no-one was recruited. Even less success was experienced through field recruitment. The MSM community, being cohesive, was easily accessed by the recruiters but did not yield many participants. Beyond this subgroup, it was clear that outside of the treatment setting, substance abusers in recovery were not a cohesive group and were not readily accessed in the field. Recruiters doing field work were more likely to make contact with persons who were actively using drugs, but no-one was referred to rehabilitation and no participant was therefore recruited through this route.
The difficulty faced in recruiting HIV-positive substance abusers in recovery is related to 2 main factors: (1) the size of the target pool was small because of high rates of relapse to drug use and low rates of HIV testing in this population, and (2) high levels of stigma associated with drug use and HIV-positive status. At all levels of the intervention, relapse to drug use contributed to a shrinking of the available pool of persons. Persons identified by clinical staff relapsed to drug use before recruitment. Even when HIV-positive substance abusers in recovery had been recruited to the project and enrolled to participate, relapse occurred before assessments were completed, and at times, before beginning any intervention sessions. Among those who completed the intervention, relapse also precluded adequate follow-up since persons relapsing were not easily contacted either directly or through their contact persons. This gave credence to the theme identified in the focus group discussions that active drug users, especially those who were aware of an HIV-positive status, were less likely to sustain abstinence or be readily accessible. Even when recruiters made contact with these individuals, there was no motivation to stop drug use.
Both crack cocaine use and HIV seropositivity are stigmatized in the Trinidad and Tobago society, especially among women. This contributes to low rates of HIV testing, nontreatment-seeking behavior, and reluctance to identify with HIV treatment interventions. Substance abusers in recovery approached for recruitment did not know their status, were HIV negative, or chose not to reveal an HIV-positive status. The relevance of stigma to low recruitment rates in this project is exemplified by the experience of a recruiter who approached a group of substance-abusing women. Disclosures of HIV status led to discomfort among group members and fears related to confidentiality, resulting in the group deciding not to participate in the project.
From the outset, the project team recognized the need for close supervision and motivation of recruiters, given the anticipated challenges. The parent project team therefore provided a recruitment officer for direct supervision of this crucial aspect of the project. He was responsible for recruitment training and one subsequent site visit was conducted to provide hands-on assistance to a flagging recruitment rate. Recruiters initially embarked on the agreed-upon strategy and liaised with clinical staff in HIV and substance abuse treatment centers. But from the outset there was a low rate of return for recruiter time invested in the project. This seemed to be the initial factor that later led to a cycle of demotivation, disillusionment, and decreasing output among the recruiters.
In weekly motivational meetings held face to face with the project coordinator and via teleconference with the recruitment officer, the recruitment approach was constantly being monitored and adjusted as the reasons for falling short became apparent. More recruiters were brought on board from the pool of trained persons and the general tone of the recruitment approach became more community outreach including the use of technology. This did not bring the required turn around that was sought. The elusive stigmatized population was not better accessed through broad-based marketing outreach at pharmacies or medical service business areas. Recruitment through party promotion and other Web sites accessed individuals who did not meet basic project requirements for participation and further drained the depleted human resource. Efforts were also made to improve recruitment rates by broadening the criteria for inclusion in the study. Once the required institutional review board approval was obtained, the abstinence requirements were made less stringent. Now the inclusion criteria were self-report of lifetime use of drugs, or self-report of binge drinking (5 or more drinks on one occasion) of alcohol in the past 3 years, or self-report of being directly affected by a primary partner's or close family member's abuse of drugs or alcohol.
As time progressed and recruitment rates continued to fall behind, various incentives failed to work. Personal demands on their time eroded commitment, and excuses from recruiters paralleled the pressure and reflected their frustrations. Changes to the payment pattern meant to evoke a greater degree of industry were ineffective. Attempts were then made to identify a recruitment coordinator locally and to turn the focus back to the treatment services as the main source of potential participants. Both proved futile. No suitable person was recognized as a recruitment coordinator and the treatment staff at the various facilities had already lost their initial zeal and enthusiasm, especially those who were directly involved with the project. Recruiters could not live up to a request for one recruited person per week but as project deadlines approached, the demand was increased to 3 to 4 per week. Inadequate communication led to rescheduling of assessments. By this time, even postintervention assessments were affected. Participants who had completed the intervention either could not be found or did not appear for scheduled assessments. Recruiters were not as assertive in tracking persons for assessments as they could have been. The responsibility fell to the more committed assessors and was eventually officially designated to the project coordinator. The demands of the scientific rigor of the project also contributed to rescheduling and failed appointments as 2 of the assessors in South Trinidad were unable to do assessments while awaiting booster training for quality control, even though these assessors had previously done assessments. Participants originally scheduled for assessments in South Trinidad therefore had to be rescheduled in the north. Apart from the inconvenience to the participants, the change cause significant time delay as recruiters, participants, and assessors had to once again find time mutually convenient to all.
The project targeted 64 participants, half of whom would be female and half of whom would be assigned to a health promotion control group. Following the onset of recruitment in September 2005, by May 2006 only 24 participants had been recruited, assessed, and were suitable for intervention. Of these, 15 completed the intervention in 2 cohorts. The first cohort of 7 males began in October 2005. The second cohort of 8 males began in May 2006. Five women were due to start a cohort in May 2006 as well but just prior to the projected start date, 2 of these fell ill needing hospitalization. Due to these challenges, in May 2006 a phase-out plan was requested. Thus the difficulties posed with recruiting sufficient participants to meet the required sample size had disrupted the timetable, preoccupied the research team, and ultimately resulted in the pilot intervention being abandoned and funding withdrawn.
Reflections
In reflecting on the benefits and costs of this pilot project, it is accepted that replication of a treatment method involves significant time and cost. Cultural adaptation, training, preparation of material, intense supervision, and sustained quality assurance come at a high cost. The pilot implementation of the CBSM concluded that Trinidad was unsuitable as a venue for a larger full-scale implementation and evaluation. Despite the inability to confirm the utility and effectiveness of the CBSM in the Trinidad population, significant knowledge and experience were gained from the project: 1. The infrastructure exists for HIV research Organizational structure and climate are receptive and appropriate. Engagement of potential beneficiaries has been demonstrated. Knowledge and expertise to implement have been proven. HIV providers can come together for the good of their patients. A manualized intervention such as the CBSM for HIV-positive substance abusers is feasible to implement. 2. Benefits accrued to the community of HIV-positive substance abusers in recovery
At exit interviews during the phase out of project, each participant indicated a real need to have an ongoing program such as the one they experienced, on a regular basis. They each said the exercises had been very useful and in some form or fashion they managed to use what they had learnt on a daily basis.
Since the dissolution of the project, requests have been made for the use of this intervention however, the commitment to the time and resource demands for training have not been forthcoming. Even though there still exists a need to scientifically document the effectiveness of this intervention, the stage has been set for implementation and evaluation under less rigorous time constraints. Solutions are still needed for the main challenges identified that is access/recruitment, external environment (stigma, financing). A culturally appropriate version of the intervention awaits validation in Trinidad and Tobago. Persons are trained in its delivery and assessment of participants. 3. Research demands versus cultural realities
In the future, greater attention should be paid to the challenges of conducting research in new cultural contexts. The rather strict demands of research will often be in opposition to cultural realities on the ground, hence the importance of building flexibility into the research protocol to respond to changing realities in the field.
The abruptness of the termination, while understandable from the perspective of project management and cost-effectiveness, was extremely difficult for those persons who had devoted their time and passion to the project.
Participants who had been screened and anticipating participation when the project ended had to be told that the project had ended. This may make it even more difficult in the future to recruit similar participants.
Recommendations for Future Interventions
The implementation of an acculturated adaptation of the CBSM in Trinidad and Tobago highlighted the importance of buy-in and stakeholder involvement in the adaptation and implementation processes. The support that was given to the project was directly related to the involvement of the various stakeholders concerned with substance abuse and HIV management. The value of input from these stakeholders through the focus groups was apparent through the accuracy of the anticipated challenges especially in the area of recruitment. A wider catchment of stakeholders to include especially the target population of HIV-positive substance abusers in recovery may have offered even more potential solutions.
The challenges associated with the research demands might be addressed by deemphasizing the research process without compromising the integrity of the research. This may be best achieved by integrating the intervention into existing treatment services. Accomplishing this presents an opportunity for the scale-up of integrated service provision to strengthen substance abuse and HIV linkages. Specific areas would require attention. These include a measure of prevalence of alcohol and drug use disorders, as well as an assessment of prevalence of the dual diagnosis of substance abuse and HIV. For the intervention to be successfully implemented would also require the promotion of HIV testing among substance abusers. Formal links would need to be established between substance abuse rehabilitation and HIV treatment services. This is likely to lead to improved access to and acceptance of the need for both services by persons dually afflicted. Staff training on the linkages between substance abuse and HIV would emphasize substance abuse treatment as a dimension of HIV prevention and increase the probability that substance abusers diagnosed with HIV will be referred for rehabilitation, and substance abusers in rehabilitation, if found HIV positive, would be referred to appropriate HIV treatment services.
Policy linkages may address the importance of mandatory HIV testing for substance abusers and motivational counseling for substance abusers diagnosed with HIV infection. An intervention such as the CBSM would be well received in such a setting of integrated service and would serve to enhance the effectiveness of both substance abuse rehabilitation and HIV treatment among the dually diagnosed population. Staff training will be needed for its implementation and policies put in place to ensure adherence to the CBSM manual. The resulting improved quality of care would no doubt improve service provider motivation and enhance cooperation with the rigors of scientific evaluation. There would also be the added benefit of reduced HIV stigmatization through the gateway of increased HIV testing. This process is likely to be hindered however by lack of human and financial resources, inadequate infrastructure, and generally poor access to the target population.
Footnotes
The author(s) declared no conflicts of interest with respect to the research, authorship, and/or publication of this article.
The Project “Cognitive Behavioural Treatment for HIV-Positive Recovering Drug Abusers” in Trinidad was supported by a supplemental grant to the National Institute of Drug Abuse, parent grant #R01 DA13802.