Abstract
Objectives. We compared Black West Indian immigrants' and US-born Blacks' sexual and drug-use risk behaviors and their beliefs related to using condoms and informing partners of sexually transmitted infections (STIs) to identify possible differences in risk.
Methods. We drew data from the baseline assessment of a clinic-based intervention designed to increase partner STI notification.
Results. Black West Indian men were less likely than were US-born Black men to report nonregular partners. There were no differences in condom use. US-born Black women were more likely than were Black West Indian women to be extremely confident that they could convince their regular partners to use condoms (odds ratio [OR] = 2.40; 95% confidence interval [CI] = 1.21, 4.76), whereas there were no differences between Black West Indian and US-born Black men on this measure (interaction P = .06). US-born Black women were more likely than were Black West Indian women to be extremely confident in their ability to discuss STI screening with their regular partners (OR = 1.89; 95% CI = 1.03, 3.47).
Conclusions. Black West Indian women's lower levels of confidence that they can discuss STI screening with their regular partners and convince these partners to use condoms may increase their infection risk. Gender-sensitive interventions are warranted for Black West Indian immigrants, especially women.
Black West Indian immigrants (hereafter West Indian immigrants) from the countries of the English-speaking Caribbean basin1,2 represent more than 20% of New York City's foreign-born population and more than one quarter of its Black population.3,4 Among Black New Yorkers newly diagnosed with HIV/AIDS, 23% are foreign born, and of these individuals, 50% originate from the Caribbean.5 Despite the size and significance of this immigrant group, little is known about these individuals' HIV and sexually transmitted infection (STI) risk and protective behaviors, which may differ considerably from those of US-born populations. If effective sexual health interventions are to be targeted toward Black populations of different national origins, a better understanding of the sexual and drug-use risk behaviors of these groups is necessary.
Characteristics associated with West Indians' migration experiences, including conditions in their countries of origin, may increase this immigrant group's risk of HIV and other STIs. The Caribbean region has the highest HIV prevalence in the Americas, approximately 2% among adults.6,7 Transnational ties among West Indian immigrants are strong, and circular migration and bidirectional travel, which may be associated with sexual mixing, are common.8,9
Gender inequalities, which are prominent in West Indian societies, may promote conditions of risk; traditional relationship norms support early sexual initiation and multiple partnerships for men but restrict power in sexual relationships for women.10–12 Violence against women is socially tolerated and only recently has been proscribed in the legal system.13–15 Open discussion of sexual matters between adults and children and between men and women is discouraged,16 and homosexuality is highly stigmatized.12
By contrast, other factors may decrease West Indian immigrants' risk of STIs, although evidence regarding such factors is limited. As an example, West Indian women frequently migrate on their own, and they participate in the workforce in high numbers in both their home countries and the United States.10,17 Their level of participation is especially high relative to other immigrant women in New York City.3 Thus, because economic independence may foster greater power in sexual relationships,18 West Indian women's higher workforce participation may increase their level of control in their sexual relationships with men. Alternately, undocumented, single immigrant women who are employed in personal and household service positions may be particularly dependent on men and may have limited power in their sexual relationships.
In addition, West Indian immigrants rank higher on various economic indicators than does the US-born New York City population as a whole.3 It has been shown (in comparisons unadjusted for age) that West Indians are more likely to be homeowners and employed and less likely to be living in poverty than US-born New Yorkers; by contrast, their median incomes are lower and they are more likely to live in women-headed households.3
Although many West Indian immigrants prosper economically after their arrival in New York City, in other cases, these immigrants' life trajectories are influenced by the same social factors that affect the experiences of US-born Blacks. Examples of such factors are exposure to racial discrimination and to the potential for social and economic marginalization,19,20 with their attendant risk of engaging in sexual acts for money or other needs, intimate partner violence, and drug and alcohol abuse.21,22 Reflective of these processes is the fact that West Indian immigrants in New York City often live in racially segregated Black neighborhoods19 and in neighborhoods in which HIV prevalence rates are among the highest in the city.23
The goal of our study was to determine whether West Indian immigrant and US-born Black STI clinic attendees differed with respect to their sexual and drug-use risk behaviors and their self-efficacy, attitudes, and intentions related to condom use and partner STI notification. We hypothesized that West Indian women would have fewer casual or 1-time partners but would less frequently engage in protected sexual intercourse than would US-born Black women and that there would be no differences in sexual risk behaviors between US-born Black men and West Indian men.
In addition, because West Indian norms discourage open discussion of sex between men and women, we expected that West Indians would not be as confident as US-born Blacks in their ability to discuss STI screening and condom use with their partners and would have less favorable attitudes toward doing so. We predicted that longer length of residence in the United States and younger age at arrival in the United States would increase West Indians' comfort in discussing these issues. We did not expect differences in drug-use behaviors between the 2 groups.
METHODS
Participants
Data were drawn from the baseline assessment of a randomized study, conducted between January 2002 and December 2004, examining the effectiveness of an intervention intended to improve STI partner notification (the study has been described in detail elsewhere24). The sample from this study provided a unique opportunity to gain an understanding of the sexual and drug-use risk behaviors and beliefs of West Indian immigrants recruited from 2 public STI clinics in Brooklyn, NY, and to compare these behaviors with those of US-born Blacks recruited from the same clinics.
Potentially eligible participants were 18 years or older, had been sexually active within the preceding 2 months, and had been referred by a provider as having probable Chlamydia trachomatis or Neisseria gonorrhoeae genital infection. A total of 783 participants met these criteria. We compared in our analyses 587 participants who, in addition to self-identifying as Black, identified themselves either (1) as West Indian or Caribbean and as being foreign born or (2) as African American and as being US born.
Measures
Trained interviewers used standardized measures to conduct structured in-person assessments in the following areas: sexual risk behaviors; self-efficacy, attitudes, and intentions related to condom use and informing partners of STIs; and substance use.
Sexual risk behaviors.
Participants' sexual behaviors in the preceding 90 days were assessed in terms of total number and gender of partners; whether they had had any regular (“someone with whom you have had an ongoing relationship and who you have sex with often—like a spouse, lover, or boyfriend or girlfriend”), casual (“someone with whom you have sex occasionally”), or 1-time (“someone you have had sex with 1 time and whom you don't plan to have sex with again”) sexual partners; and frequency of condom use during vaginal and anal intercourse with each type of partner (always, almost always, sometimes, or never; dichotomized into never vs at least sometimes). Participants were also asked to provide detailed information, including gender and length of relationship, on up to 5 of their most recent sexual partners (in the preceding 90 days).
Self-efficacy, attitudes, and intentions.
Participants were asked to indicate, for each type of partner reported (regular, casual, 1 time), their overall level of confidence that in the next 90 days they could convince these partners to use a condom, the extent to which they felt favorably or unfavorably about always using a condom with these partners, and how likely or unlikely it was that they would always use a condom with these partners.25 Responses were made on 7-point scales ranging from extremely confident, favorable, or likely to extremely unconfident, unfavorable, or unlikely.
Partner notification beliefs were ascertained separately for each of the participants' 5 most recent partners. Participants were asked to imagine that their doctor or other health care provider wanted them to tell this partner that he or she needed to undergo STI screening and to indicate (1) how confident they were that they could bring up the topic, (2) how confident they were that they could convince this partner to undergo STI screening, (3) the extent to which they felt favorably or unfavorably about talking to this partner about being screened, and (4) the likelihood that they would notify this partner about STI screening. Again, responses were made on 7-point scales ranging from extremely confident, favorable, or likely to extremely unconfident, unfavorable, or unlikely. For each participant, we calculated a mean value across all reported partners of each type (regular, casual, 1 time).
Because self-efficacy, attitude, and intention scores for both condom use and STI partner notification were highly skewed, we used the Mann–Whitney test, stratified by gender, to compare US-born Blacks and West Indians. Then we dichotomized responses into “extremely” (i.e., a score of 7) versus all other categories and conducted the χ2 test to assess associations. Comparisons from the Mann–Whitney test were consistent with comparisons from the χ2 test. We report results from analyses in which the dichotomized variables were used (i.e., “extremely” vs all other categories).
Substance use.
Alcohol use was measured with the Alcohol Use Disorders Identification Test (AUDIT).26 Participants were asked to indicate how often they consumed alcohol and how often they had experienced problems related to alcohol use. AUDIT scores of 8 and above indicate a likelihood of harmful alcohol consumption. Participants' history of use of marijuana and other illegal drugs (e.g., heroin, crack, cocaine, methamphetamines, inhalants, street methadone, any injected drug) was ascertained, as well as their frequency of use of these drugs over the preceding 90 days.
Statistical Analyses
We used the χ2 test, the Fisher exact test, the t test, and the Mann–Whitney test, as appropriate (with the α level set at .05), to compare West Indian and US-born Black women and men with respect to demographic characteristics and sexual and drug-use risk behaviors. We used linear regression (for continuous measures) and binary logistic regression (for dichotomous measures) to assess main effects of ethnicity (US-born Black vs West Indian), after we controlled for gender, on demographic characteristics, risk behaviors, and condom use and partner notification variables, as well as to assess interactions between gender and ethnicity and any of these factors. (α levels were set at .05 for main effects and .10 for interactions). We evaluated whether mean length of relationship (for regular and casual partners only), educational level, and income level altered the associations between ethnicity and each outcome.
In analyses among West Indians only, we used binary logistic regression to assess the effects of length of residence in the United States and age at arrival on self-efficacy, attitudes, and intentions related to condom use and informing partners of STIs. Length of residence in the United States was categorized as less than 5 years, 5 to 9 years, or 10 or more years; age at arrival was categorized as less than 12 years, 12 to 18 years, 19 to 24 years, or 25 years or older. We also estimated models including uncategorized measures of length of residence in the United States and age at arrival.
RESULTS
Table 1 shows the demographic characteristics of the 2 groups, stratified according to gender. The mean age of West Indian and US-born Black participants did not differ. Educational levels were significantly lower among West Indian women and men than among US-born Black women and men. A higher percentage of West Indian women than US-born Black women was living with a partner (33.1% vs 22.0%; P = .06). There were no significant interactions between ethnicity, gender, and the demographic variables.
TABLE 1.
Demographic Characteristics of Black West Indian Immigrants and US-Born Blacks Attending Sexually Transmitted Infection (STI) Clinics: New York City, January 2002–December 2004
| Black Women |
Black Men |
|||||
| West Indian | US Born | Pa | West Indian | US Born | Pa | |
| Total, No. | 136 | 100 | 151 | 200 | ||
| Age, y, mean (SD) | 24.5 (5.16) | 24.8 (6.67) | .70 | 28.0 (7.69) | 28.1 (8.19) | .99 |
| Income, $, No. (%) | .32 | .54 | ||||
| <18 000 | 103 (83.1) | 71 (74.7) | 82 (56.9) | 98 (53.8) | ||
| 18 000–36 000 | 17 (13.7) | 19 (20.0) | 48 (33.3) | 59 (32.4) | ||
| >36 000 | 4 (3.2) | 5 (5.3) | 14 (9.7) | 25 (13.7) | ||
| Highest education level, No. (%) | .04 | <.01 | ||||
| Junior high school (8th grade) or less | 7 (5.1) | 1 (1.0) | 10 (6.6) | 2 (1.0) | ||
| High school diploma or less | 104 (76.5) | 69 (69.0) | 109 (72.2) | 129 (64.8) | ||
| Some college or college degree | 25 (18.4) | 30 (30.0) | 32 (21.2) | 68 (34.2) | ||
| No. of people in household, mean (SD) | 2.3 (1.4) | 2.2 (1.4) | .68 | 2.3 (1.6) | 2.1 (1.4) | .29 |
| Living with partner or spouse, No. (%) | 45 (33.1) | 22 (22.0) | .06 | 42 (27.8) | 48 (24.1) | .43 |
| Length of residence in the United States, y, No. (%) | .11b | |||||
| <1 | 6 (4.4) | 4 (2.7) | ||||
| 1–4 | 48 (35.3) | 36 (24.0) | ||||
| 5–9 | 34 (25.0) | 40 (26.7) | ||||
| ≥10 | 48 (35.3) | 70 (46.7) | ||||
Note. There were no significant interactions between gender or ethnicity and any demographic variable.
Within-gender comparison of ethnic groups (Pearson χ2 test or t test).
Comparison of Black West Indian men and women (χ2 test).
Risk Behaviors
Table 2 presents a comparison of West Indian and US-born Black men and women with respect to sexual and drug-use risk behaviors. Substantial percentages (above 40%) of both groups reported a history of STIs. West Indian men were less likely than were US-born Black men to report casual (43.0% vs 55.8%; P = .02) and 1-time (32.5% vs 43.0%; P = .04) partners, and West Indian women were less likely than were US-born Black women to report 1-time partners (8.8% vs 18.0%; P = .04).
TABLE 2.
Sexual Risk Behaviors and Drug and Alcohol Use Among Black West Indian Immigrants and US-Born Blacks Attending Sexually Transmitted Infection (STI) Clinics: New York City, January 2002–December 2004
| Black Women |
Black Men |
|||||
| West Indian | US Born | Pa | West Indian | US Born | Pa | |
| Total, No. | 136 | 100 | 151 | 200 | ||
| STI history, No. (%) | ||||||
| Ever had an STI | 59 (43.4) | 54 (54.0) | .11 | 65 (43.0) | 87 (43.5) | .93 |
| Underwent an HIV test | 125 (91.9) | 91 (91.0) | .80 | 132 (87.4) | 173 (86.5) | .80 |
| HIV positiveb | 0 (0.0) | 2 (3.1) | .15c | 2 (1.9) | 5 (4.0) | .46c |
| Sexual behavior in past 90 days | ||||||
| 5 or more partners, No. (%) | 0 (0) | 4 (4.0) | .03c | 13 (8.6) | 29 (14.5) | .10 |
| No. of partners in past 90 days, median (range) | 1 (1–4) | 1 (1–7) | .22 | 2 (1–10) | 2 (1–21) | .02 |
| Had regular partner(s), No. (%) | 126 (92.6) | 89 (89.0) | .33 | 128 (84.8) | 156 (78.0) | .11 |
| Length of relationship with regular partner(s), mo, mean (SD) | 35.4 (47.2) | 38.3 (88.6) | .76 | 37.0 (50.9) | 38.5 (69.2) | .85 |
| Had vaginal intercourse with regular partner(s), No. (%) | 126 (100.0) | 87 (97.8) | .17c | 126 (98.4) | 151 (96.8) | .46c |
| Frequency of condom use during vaginal intercourse with regular partner(s), No. (%) | .77 | .77 | ||||
| Never | 64 (50.8) | 46 (52.9) | 59 (46.8) | 68 (45.0) | ||
| At least sometimes | 62 (49.2) | 41 (47.1) | 67 (53.2) | 83 (55.0) | ||
| Had casual partner(s), No. (%) | 27 (19.9) | 29 (29.0) | .10 | 65 (43.0) | 111 (55.8) | .02 |
| Length of relationship with casual partner(s), mo, mean (SD) | 11.0 (47.2) | 20.9 (27.7) | .10 | 13.9 (18.5) | 19.8 (30.8) | .11 |
| Had vaginal intercourse with casual partner(s), No. (%) | 26 (96.3) | 27 (93.1) | ≥.99c | 64 (98.5) | 103 (92.8) | .16c |
| Frequency of condom use during vaginal intercourse with casual partner(s), No. (%) | .22 | .71 | ||||
| Never | 4 (15.4) | 8 (29.6) | 12 (18.8) | 17 (16.5) | ||
| At least sometimes | 22 (84.6) | 19 (70.4) | 52 (81.3) | 86 (83.5) | ||
| Had 1-time partner(s), No. (%) | 12 (8.8) | 18 (18.0) | .04 | 49 (32.5) | 86 (43.0) | .04 |
| Had vaginal intercourse with 1-time partner(s), No. (%) | 12 (100.0) | 15 (83.3) | .26c | 45 (91.8) | 70 (81.4) | .10 |
| Frequency of condom use during vaginal intercourse with 1-time partner(s), No. (%)d | ≥.99c | .11 | ||||
| Never | 7 (58.3) | 9 (60.0) | 30 (66.7) | 56 (80.0) | ||
| At least sometimes | 5 (41.7) | 6 (40.0) | 15 (33.3) | 14 (20.0) | ||
| Had anal intercourse with any partner, No. (%) | 7 (5.1) | 12 (12.0) | .06 | 22 (14.6) | 51 (25.5) | .01 |
| Frequency of condom use during anal intercourse with any partner(s), No. (%) | .02c | .37 | ||||
| Never | 1 (14.3) | 9 (75.0) | 10 (45.5) | 29 (56.9) | ||
| At least sometimes | 6 (85.7) | 3 (25.0) | 12 (54.5) | 22 (43.1) | ||
| Asked a partner to use condom, No. (%) | 82 (60.3) | 51 (51.0) | .16 | 60 (39.7) | 86 (43.2) | .51 |
| Partner(s) refused to use condom when asked, No. (%) | 33 (40.2) | 16 (31.4) | .30 | 14 (23.3) | 21 (24.4) | .88 |
| Decided not to have sex because condom was not available or partner did not want to use a condom, No. (%)e | 31 (22.8) | 23 (23.0) | .97 | 39 (25.8) | 79 (39.9) | .01 |
| Given or received sexual acts for drugs, food, or money, No. (%) | 0 (0.0) | 3 (3.0) | .08c | 5 (3.3) | 10 (5.0) | .43 |
| Drug- and alcohol-use behaviors | ||||||
| Used marijuana in past 90 days, No. (%) | 30 (54.5) | 21 (51.2) | .75 | 75 (74.3) | 112 (73.7) | .92 |
| Used drugs other than marijuana in past 90 days, No. (%) | 0 (0.0) | 1 (1.0) | .42c | 1 (0.7) | 11 (5.5) | .01 |
| Alcoholism scale score, mean (SD) | 1.83 (1.5) | 2.37 (2.7) | .19 | 3.21 (3.1) | 3.92 (3.6) | .13 |
| Alcohol use frequency, No. (%) | .41 | .14 | ||||
| Never | 78 (57.8) | 50 (50.0) | 49 (32.5) | 61 (30.5) | ||
| 1–3 times monthly | 45 (33.3) | 37 (37.0) | 63 (41.7) | 72 (36.0) | ||
| Once a week | 9 (6.7) | 9 (9.0) | 25 (16.6) | 27 (13.5) | ||
| Several times weekly | 3 (2.2) | 2 (2.0) | 11 (7.3) | 29 (14.5) | ||
| Daily | 0 (0.0) | 2 (2.0) | 11 (5.5) | |||
Comparison of ethnic groups, stratified by gender (χ2 test, Mann–Whitney test, t test, or Fisher exact test, as appropriate).
As a result of missing data, sample sizes were 65, 103, 126, and 103.
From Fisher exact test because of small sample sizes.
Significant interaction (P = .09) between gender and ethnicity.
As a result of missing data, sample sizes were 198 and 151 for men.
There were no significant differences between West Indian and US-born Black women and men in reported frequency of condom use during vaginal intercourse with any type of partner or in reports of partners refusing to use condoms; however, US-born Black men were more likely than were West Indian men to report that they had decided to forgo sexual intercourse because a condom was not available. West Indians were less likely than were US-born Blacks to report anal intercourse, even when same-gender encounters were excluded (data not shown). West Indian women were more likely than were US-born Black women to report using condoms during anal intercourse, whereas there were no differences between the 2 groups of men (interaction P = .09).
West Indian men were less likely than were US-born Black men to have ever used drugs other than marijuana (Table 2). There were no differences between West Indians and US-born Blacks with respect to reported frequency of alcohol or marijuana use or mean AUDIT score.
Controls for income and education did not alter any of the results shown in Table 2 with the exception of the association between ethnicity and engaging in sexual acts for drugs, food, or money. The adjusted odds ratio (OR) for engaging in sexual acts for drugs, food, or money among US-born Blacks (vs West Indians) was 4.31 (95% confidence interval [CI] = 1.13, 16.44).
Condom Use Self-Efficacy, Attitudes, and Intentions
Table 3 displays results for self-efficacy, attitudes, and intentions related to condom use, with each measure treated as an outcome. Shown are the percentages of participants scoring extremely high on each measure for each partner type, by gender and ethnicity. ORs relating ethnicity to outcomes, after control for gender, are shown if no interaction was observed. When the interaction between gender, ethnicity, and the outcome measure was significant, ORs for ethnicity are reported separately for men and women. Models included controls for relationship length when that variable altered the estimate. Because the addition of income level and educational level did not materially alter the results, models did not include controls for these variables.
TABLE 3.
Self-Efficacy, Attitudes, and Intentions Relating to Condom Use With Partners Among Black West Indian Immigrants and US-Born Blacks Attending Sexually Transmitted Infection Clinics: New York City, January 2002–December 2004
| Women |
Men |
Overall Ethnicity,b ORe (95% CI) | Interaction P | Ethnicityc |
||||
| No. (%) | Pa | No. (%) | Pa | Women, OR (95% CI) | Men, OR (95% CI) | |||
| Regular partnersd | ||||||||
| Extremely confident in ability to convince partners to use condoms | .01 | .94 | .06 | |||||
| Black West Indians | 87 (69.0) | 98 (76.6) | Reference | Reference | ||||
| US-born Blacks | 75 (84.3) | 120 (76.9) | 2.40 (1.21, 4.76) | 1.02 (0.59, 1.77) | ||||
| Extremely favorable toward always using condoms with partners | .02 | .31 | .21 | |||||
| Black West Indians | 83 (65.9) | 83 (64.8) | Reference | |||||
| US-born Blacks | 72 (80.9) | 110 (70.5) | 1.59 (1.07, 2.35) | |||||
| Extremely likely to always use condoms with partners | .42 | .42 | .30 | |||||
| Black West Indians | 80 (63.5) | 82 (64.1) | Reference | |||||
| US-born Blacks | 68 (76.4) | 107 (68.6) | 1.45 (0.99, 2.12) | |||||
| Casual partnerse | ||||||||
| Extremely confident in ability to convince partners to use condoms | .27 | .70 | .26 | |||||
| Black West Indians | 25 (92.6) | 58 (89.2) | Reference | |||||
| US-born Blacks | 24 (82.8) | 101 (91.0) | 0.88 (0.37, 2.11) | |||||
| Extremely favorable toward always using condoms with partners | .16 | .80 | .27 | |||||
| Black West Indians | 25 (92.6) | 59 (90.8) | Reference | |||||
| US-born Blacks | 23 (79.3) | 102 (91.9) | 0.75 (0.30, 1.86) | |||||
| Extremely likely to always use condoms with partners | .01 | .69 | .07 | |||||
| Black West Indians | 25 (92.6) | 57 (87.7) | Reference | Reference | ||||
| US-born Blacks | 19 (65.5) | 95 (85.6) | 0.15 (0.03, 0.78) | 0.83 (0.34, 2.07) | ||||
| One-time partnersf | ||||||||
| Extremely confident in ability to convince partners to use condoms | .66 | .10 | .20 | |||||
| Black West Indians | 10 (83.3) | 47 (95.9) | Reference | |||||
| US-born Blacks | 16 (88.9) | 75 (87.2) | 0.49 (0.15, 1.57) | |||||
| Extremely favorable toward always using condoms with partners | .66 | .91 | .71 | |||||
| Black West Indians | 10 (83.3) | 48 (98.0) | Reference | |||||
| US-born Blacks | 16 (88.9) | 84 (97.7) | 1.22 (0.26, 5.88) | |||||
| Extremely likely to always use condoms with partners | .58 | .44 | .35 | |||||
| Black West Indians | 9 (75.0) | 48 (98.0) | Reference | |||||
| US-born Blacks | 15 (83.3) | 82 (95.3) | 0.91 (0.24, 3.38) | |||||
Note. OR = odds ratio; CI = confidence interval.
Between-group comparison (χ2).
Odds of being extremely confident, favorable, or likely versus not confident, favorable, or likely in comparisons of US-born Blacks (coded 1) and Black West Indians (coded 0) after we had controlled for gender. Each row represents a separate model. Estimates are not shown in cases of a significant interaction.
ORs are shown when the interaction between gender, ethnicity, and the outcome measure was significant at P ≤ .10.
Sample sizes were as follows: 89 US-born Black women, 126 Black West Indian women, 156 US-born Black men, 128 Black West Indian men.
Sample sizes were as follows: 29 US-born Black women, 27 Black West Indian women, 111 US-born Black men, 65 Black West Indian men.
Sample sizes were as follows: 18 US-born Black women, 12 Black West Indian women, 86 US-born Black men, 49 Black West Indian men.
US-born Blacks were more likely than were West Indians to be extremely favorable toward always using condoms with their regular partners in the subsequent 90 days (OR = 1.59; 95% CI = 1.07, 2.35) and to report that they would be extremely likely to always do so (OR = 1.45; 95% CI = 0.99, 2.12; Table 3). A significant interaction was found between gender, ethnicity, and confidence in one's ability to convince one's regular partners to use condoms. This interaction indicated that US-born Black women were more likely than were West Indian women to be extremely confident in their ability to convince their partners to use a condom (OR = 2.40; 95% CI = 1.21, 4.76), whereas there were no differences between the 2 groups of men (OR = 1.02; 95% CI = 0.59, 1.77; interaction P = .06). Controls for mean length of relationship did not alter these associations.
In contrast to the findings for regular partners, US-born Black women were less likely than were West Indian women to report being extremely likely to always use condoms with their casual partners in the subsequent 90 days (OR = 0.15; 95% CI = 0.03, 0.78), a difference not found among men (interaction P = .07; Table 3). There were no significant differences in the case of 1-time partners.
Partner Notification Self-Efficacy, Attitudes, and Intentions
Table 4 displays results for self-efficacy, attitudes, and intentions related to partner notification of an STI. US-born Blacks were more likely than were West Indians to report being extremely likely to inform their regular partners that they had an STI (OR = 1.60; 95% CI = 0.96, 2.63; Table 4). US-born Black women were more likely than were West Indian women to be extremely confident that they could discuss STI screening with their regular partners (OR = 1.89, 95% CI = 1.03, 3.47); there were no differences between the 2 groups of men (interaction P = .10).
TABLE 4.
Self-Efficacy, Attitudes, and Intentions Relating to Partner Notification Among Black West Indian Immigrants and US-Born Blacks Attending Sexually Transmitted Infection (STI) Clinics: New York City, January 2002–December 2004
| Women |
Men |
Overall Ethnicity, ORb (95% CI) | Interaction P | Ethnicityc |
||||
| No. (%) | Pa | No. (%) | Pa | Women, OR (95% CI) | Men, OR (95% CI) | |||
| Regular partnersd | ||||||||
| Extremely confident in ability to discuss STI screening with partners | .17 | .04 | .02 | |||||
| Black West Indians | 92 (73.0) | 105 (82.7) | Reference | Reference | ||||
| US-born Blacks | 73 (81.1) | 111 (72.1) | 1.59 (0.82, 3.07) | 0.54 (0.30, 0.97) | ||||
| Extremely confident in ability to convince partners to undergo | .04 | .71 | .10 | |||||
| STI screening | ||||||||
| Black West Indians | 80 (63.5) | 110 (86.6) | Reference | Reference | ||||
| US-born Blacks | 69 (76.7) | 131 (85.1) | 1.89 (1.03, 3.47) | 0.88 (0.45, 1.73) | ||||
| Extremely favorable toward discussing STI screening with partners | .15 | .16 | .05 | |||||
| Black West Indians | 90 (71.4) | 111 (87.4) | Reference | Reference | ||||
| US-born Blacks | 72 (80.0) | 125 (81.2) | 1.60 (0.84, 3.05) | 0.62 (0.32, 1.20) | ||||
| Extremely likely to notify partners to undergo STI screening | .05 | .55 | .35 | |||||
| Black West Indians | 94 (74.6) | 110 (86.6) | Reference | |||||
| US-born Blacks | 77 (85.6) | 137 (89.0) | 1.60 (0.97, 2.63) | |||||
| Casual partnerse | ||||||||
| Extremely confident in ability to discuss STI screening with partners | .28 | .84 | .36 | |||||
| Black West Indians | 16 (59.3) | 45 (68.2) | Reference | |||||
| US-born Blacks | 13 (44.8) | 76 (66.7) | 0.79 (0.45, 1.38) | |||||
| Extremely confident in ability to convince partners to undergo | .06 | .61 | .21 | |||||
| STI screening | ||||||||
| Black West Indians | 16 (59.3) | 49 (74.2) | Reference | |||||
| US-born Blacks | 10 (34.5) | 80 (70.2) | 0.62 (0.35, 1.12) | |||||
| Extremely favorable toward discussing STI screening with partners | .001 | .88 | .01 | |||||
| Black West Indians | 20 (74.1) | 47 (71.2) | Reference | Reference | ||||
| US-born Blacks | 9 (31.0) | 80 (70.2) | 0.12 (0.03, 0.42) | 0.95 (0.48, 1.86) | ||||
| Extremely likely to notify partners to undergo STI screening | .26 | .40 | .12 | |||||
| Black West Indians | 18 (66.7) | 51 (77.3) | Reference | |||||
| US-born Blacks | 15 (51.7) | 94 (82.5) | 0.99 (0.53, 1.86) | |||||
| One-time partnersf | ||||||||
| Extremely confident in ability to discuss STI screening with partners | .65 | .18 | .82 | |||||
| Black West Indians | 7 (58.3) | 35 (74.5) | Reference | |||||
| US-born Blacks | 9 (50.0) | 53 (63.1) | 0.61 (0.31, 1.23) | |||||
| Extremely confident in ability to convince partners to undergo | .46 | .004 | .64 | |||||
| STI screening | ||||||||
| Black West Indians | 7 (58.3) | 38 (80.9) | Reference | |||||
| US-born Blacks | 7 (38.9) | 47 (56.0) | 0.33 (0.16. 0.69) | |||||
| Extremely favorable toward discussing STI screening with partners | .46 | .05 | .88 | |||||
| Black West Indians | 6 (50.0) | 37 (78.7) | Reference | |||||
| US-born Blacks | 6 (33.3) | 52 (61.9) | 0.45 (0.22, 0.93) | |||||
| Extremely likely to notify partners to undergo STI screening | .71 | .30 | .26 | |||||
| Black West Indians | 6 (50.0) | 38 (80.9) | Reference | |||||
| US-born Blacks | 11 (61.1) | 60 (71.4) | 0.76 (0.36, 1.57) | |||||
Note. OR = odds ratio; CI = confidence interval. Sample sizes (listed in footnotes c–e) reflect the numbers of participants responding to items on regular, casual, and 1-time partners and vary slightly from those reporting regular, casual, or 1-time partners (Table 2).
Between-group comparison (χ2).
Odds of being extremely confident, favorable, or likely versus not confident, favorable, or likely in comparisons of US-born Blacks (coded 1) and Black West Indians (coded 0) after we controlled for gender and, in casual partner models only, length of relationship. Each row represents a separate model. Estimates are not shown in cases of a significant interaction.
ORs (with control for length of relationship in casual partner models) are shown when the interaction between gender, ethnicity, and the outcome measure was significant at P ≤ .10.
Sample sizes were as follows: 90 US-born Black women, 126 Black West Indian women, 154 US-born Black men, 127 Black West Indian men.
Sample sizes were as follows: 29 US-born Black women, 27 Black West Indian women, 114 US-born Black men, 66 Black West Indian men.
Sample sizes were as follows: 18 US-born Black women, 12 Black West Indian women, 84 US-born Black men, 47 Black West Indian men.
We found similar interactions between gender, ethnicity, and (1) a high level of confidence on the part of participants that they could convince their regular partners to undergo STI screening and (2) an extremely positive attitude toward talking to their regular partners about STI screening. These interactions indicate that West Indian women were less likely than were US-born Black women to have high scores on these measures, whereas there were no differences between US-born Black men and West Indian men (Table 4).
US-born Black women were less likely than were West Indian women to have an extremely favorable attitude toward discussing STI screening with casual partners (OR = 0.12; 95% CI = 0.03, 0.42); there were no differences between men in the 2 groups (interaction P = .01 after control for length of relationship; Table 4). US-born Blacks were less likely than were West Indians to be extremely confident that they could convince 1-time partners to undergo STI screening (OR = 0.33; 95% CI = 0.16, 0.69) or to have an extremely favorable attitude toward discussing STI screening with 1-time partners (OR = 0.45; 95% CI = 0.22, 0.93); no significant interactions were found.
Finally, analyses of data from West Indian immigrants only showed that there were no significant associations between length of residence in the United States or age at arrival and any of the condom use or partner notification variables (data not shown).
DISCUSSION
A large body of research indicates that immigrants to the United States, regardless of their race or ethnicity, have an advantage over native-born Americans across diverse health behaviors and outcomes; however, this advantage declines as length of residence in the United States and degree of acculturation increase.27–33 Less is known about the HIV-related risk behaviors of immigrants than about other aspects of their mental and physical health, and most of the literature pertains to Mexican immigrants residing in the southwestern United States.21,34–37 Moreover, scant information exists on the HIV-related risk profiles of Black immigrants,32 who are indistinguishable from US-born Blacks with respect to public health indicators. In contrast to a substantial literature demonstrating that the behaviors of immigrants, including Black immigrants, are healthier than those of US-born populations, we found a mixed pattern in regard to sexual health behaviors and beliefs.
In our study, West Indians scored lower than did US-born Blacks on measures assessing key cognitions relevant to condom use. West Indians had less favorable attitudes toward using condoms with their regular partners and were less likely to intend to always use condoms with these partners in the future; such favorable attitudes toward condom use and intentions to use condoms are important predictors of actual use.38
As expected, we also found that gender had varied effects in the 2 groups. West Indian women were not as likely as US-born Black women to be confident that they could convince their regular partners to use condoms, whereas there were no differences in the confidence levels of West Indian men and US-born Black men.
We found a similar pattern with respect to whether participants would notify their partner that they had an STI, which is an important factor in STI control.39 Overall, West Indians were not as likely as were US-born Blacks to indicate that they would notify their regular partners that they had an STI. Furthermore, West Indian women had lower scores than did US-born Black women on 3 measures related to STI notification with regular partners: confidence that they could discuss STI screening with their regular partners, a favorable attitude toward discussing STI screening with their regular partners, and confidence that they could convince their regular partners to undergo STI screening.
In contrast to our finding of greater vulnerability among West Indian immigrants, especially women, in relation to self-efficacy, attitudes, and intentions to use condoms and notify regular partners of STIs, West Indians had lower risk scores than did US-born Blacks on some behavioral measures. West Indians reported fewer nonregular partners than did US-born Blacks, a finding we had expected among women but not among men, and rates of use of illicit drugs other than marijuana were lower among West Indian men than among US-born Black men.
These findings are similar to those of studies showing lower sexual and alcohol-use risk behaviors among Latino immigrants.21,34–36 Notably, however, there were no differences in rates of condom use during vaginal intercourse between West Indian and US-born Black women or men (counter to our expectation for women); rates were low among all of these STI clinic attendees.
If our participants' reports are factual, West Indian immigrants' reduced likelihood of having any casual or 1-time partners may place them at lower risk of exposure to HIV or other STIs than US-born Blacks. However, their low levels of condom use place both West Indians and US-born Blacks at risk of infection if they are exposed to an infected partner.
It might seem puzzling that US-born Blacks did not report higher levels of condom use with their regular partners, which would have been consistent with their more favorable condom-related beliefs. Although reporting bias could account for this result, a better explanation is that these beliefs were assessed in relation to future behavior, and having just been diagnosed with an STI is a powerful motivator for behavior change. Thus, West Indian men and women had less favorable attitudes toward condom use and weaker intentions to use them with regular partners in the future than did US-born Blacks in the same circumstances. Moreover, newly diagnosed West Indian women were less confident than newly diagnosed US-born Black women in their ability to convince their regular partners to use condoms.
Another puzzling finding was that West Indian immigrants scored more positively than US-born Blacks on some protective beliefs in relation to casual and 1-time partners, in contrast to their lower scores on these measures in relation to regular partners. Given that the number of women reporting nonregular partners was low (and therefore estimates might be unstable), we have less confidence in our findings for casual and 1-time partners than in our findings for regular partners.
Limitations
One limitation of this study is that our measures of attitudes and behaviors regarding condom use did not account for gender inequality in control over condom use. Although both men and women may need to convince their partners that a condom should be used in a sexual encounter, there are important differences between a man convincing his partner that a condom should be used and a woman doing so. These differences could have affected comparisons between women and men, but it is not likely that they altered the results of the comparisons of women in the 2 ethnic groups. Similarly, our failure to find differences in self-efficacy, attitudes, and intentions to use condoms and notify partners of STIs according to length of residence in the United States or age at migration may reflect the fact that these indicators of acculturation are crude.40
Because a sample of individuals with probable STI selects for higher-risk individuals, it is not possible to draw inferences from our data concerning the larger population of West Indian immigrants and US-born Blacks. It is notable, therefore, that despite the “equalizing” effect of selection of individuals who had an STI from both the West Indian and US-born Black groups, we found between-group differences in behaviors as well as perceptions related to protective behaviors.
Conclusions
Our findings suggest that (with respect to our STI clinic population) West Indian immigrants, especially men, may be at lower risk of exposure to HIV and other STIs than US-born Blacks because they are less likely to have casual or 1-time partners. Yet, West Indian immigrants, especially women, may be at higher risk of STI infection if exposed by a regular partner, because West Indian women are less confident in their ability to convince regular partners to use a condom and undergo STI screening.
Even though West Indian women may have relatively more economic power than do immigrant women from other regions, norms supporting gendered power inequalities in relationships may constrain them from talking to their regular partners about condom use and notifying their partners of their STI diagnosis and the need for testing. West Indian men's perceptions regarding these behaviors, which are less favorable in some cases than those of US-born Black men, may also influence risk in regular partnerships. Thus, in situations in which they have partners who engage in risky behaviors or do not inform them if they have an STI, West Indian women seem to lack the requisite attitudes and skills to alter patterns of risky behavior within their relationships or to respond to such behaviors.
Overall, our findings highlight significant heterogeneity among Black populations that typically is masked as a result of the use of broad racial/ethnic categories in public health research. Future studies are needed to assess the validity of these findings in representative samples and to rule out reporting bias as a possible explanation for the differences we found. In-depth research needs to focus on sociocultural and sociostructural factors, including educational level, upbringing in one's country of origin, migration experiences, and postmigration life circumstances that may explain the reluctance of West Indian immigrant women to discuss condom use and STI screening with their regular partners.
We urge public health officials and researchers to carefully assess place of birth in routine public health data and in detailed surveys of Black populations so that they can gain a better understanding of HIV- and STI-related risks among all Black immigrants. Moreover, to address the challenges that West Indian women face in talking to their regular partners about condom use and STI screening, there is an urgent need for interventions that target gender norms and behaviors among West Indian immigrants.
Acknowledgments
This research was supported by the Centers for Disease Control and Prevention (grant R30 CCR219136). S. Hoffman received support from the HIV Center for Clinical and Behavioral Studies (grant P30-MH43520), and S. T. Beckford Jarrett received support from the National Institute of Mental Health (grant T32-MH19139).
We thank the following people for their contributions to this research: Angelette Cintron, Rhonda Curney, and Joy Williams (State University of New York Downstate Medical Center); Lucindy Williams (New York City Department of Health); and Lorraine DuBouchet and ChellyAnn Hinds (Kings County Hospital Center). We also thank the staff and administrators at the Kings County Hospital Center in Brooklyn, NY; the New York City Department of Health, Bureau of STI Control; and the New York City Department of Health STI Clinic, Brooklyn. We thank Shari Dworkin, Martin Eisenberg, Jennifer Higgins, and 3 anonymous reviewers for comments on earlier versions.
Note. The findings and conclusions in this report are those of the authors and do not necessarily represent the views of the Centers for Disease Control and Prevention.
Human Participant Protection
This study was approved by the institutional review boards of the New York State Psychiatric Institute, the State University of New York Downstate Medical Center, the Kings County Hospital Center, and the New York City Department of Health. All participants provided written informed consent.
Peer Reviewed
Contributors
S. Hoffman conceptualized and oversaw analyses and wrote and edited the article. S. T. Beckford Jarrett helped conceptualize analyses and contributed to the data analysis and the writing of the article. E. A. Kelvin conducted the data analyses and contributed to the writing of the article. S. A. Wallace contributed to the writing of the article. M. Augenbraun, M. Hogben, N. Liddon, W. M. McCormack, and S. Rubin contributed substantially to the parent study conception and to the writing of the article. M. Augenbraun, W. M. McCormack, and S. Rubin contributed to study implementation. T E.. Wilson designed the parent study, oversaw all aspects of its implementation, and contributed to the writing of the article.
References
- 1.Voelker R. HIV/AIDS in the Caribbean. JAMA 2001;285:2961–2963 [DOI] [PubMed] [Google Scholar]
- 2.Kasinitz P. Caribbean New York: Black Immigrants and the Politics of Race . Ithaca, NY: Cornell University Press; 1992 [Google Scholar]
- 3. The Newest New Yorkers 2000: Immigrant New York in the New Millennium . New York, NY: New York City Dept of City Planning; 2004 [Google Scholar]
- 4. Census 2000 Supplementary Survey Profile . Washington, DC: US Census Bureau; 2000 [Google Scholar]
- 5. HIV/AIDS in New York City, 2004: Blacks . New York, NY: New York City Dept of Health and Mental Hygiene; 2006 [Google Scholar]
- 6. AIDS Surveillance in the Americas . Washington, DC: Pan American Health Organization; 2002 [Google Scholar]
- 7. 2004 Report on the Global HIV/AIDS Epidemic . Geneva, Switzerland: Joint United Nations Programme on HIV/AIDS; 2004 [Google Scholar]
- 8.Pessar PR. The role of gender, households, and social networks in the migration process: a review and appraisal. In: Hirschman C, Kasinitiz P, DeWind J, eds. The Handbook of International Migration: The American Experience New York, NY: Russell Sage Foundation; 1999:53–70 [Google Scholar]
- 9.Foner N. Transnationalism old and new. In: Foner N, ed. In a New Land New York: New York University Press; 2005:62–88 [Google Scholar]
- 10.Leo-Rhynie EA. Socialisation and the development of gender identity: theoretical formulations and Caribbean research. In: Barrow C, ed. Caribbean Portraits: Essays on Gender Ideologies and Identities Kingston, Jamaica: Ian Randle Publishers; 1998:234–252 [Google Scholar]
- 11.Brown J, Newland A, Anderson P, Chevannes B. Caribbean fatherhood: underresearched, misunderstood. In: Roopnarine JL, Brown J, eds. Caribbean Families: Diversity Among Ethnic Groups Greenwich, Conn: Ablex Publishing Co; 1997:85–113 [Google Scholar]
- 12.Chevannes B. Learning to Be a Man: Culture, Socialization and Gender Identity in Five Caribbean Communities . Kingston, Jamaica: University of the West Indies Press; 2001 [Google Scholar]
- 13.Johnson T. The Intersection of Two Global Epidemics: HIV/AIDS and Gender-Based Violence [master's thesis] . New York, NY: Columbia University; 2003 [Google Scholar]
- 14.Lazarus-Black M. The (heterosexual) regendering of a modern state: criminalizing and implementing domestic violence law in Trinidad. Law Soc Inquiry 2003;28:979–1008 [Google Scholar]
- 15.Bylander KE, Hoffman S, Beckford S, Mantell JE, Carroll T, Augenbraun M. “At home the cops don't come”: migration and changing views of relationship violence among West Indian immigrants. Presented at: 133rd Annual Meeting of the American Public Health Association; December 10–14, 2005; Philadelphia, Pa. Abstract 110300
- 16.Harris-Hastick E, Modeste-Curwen C. The importance of culture in HIV/AIDS prevention in Grenada. J HIV/AIDS Prev Educ Adolesc Child 2001;4:5–22 [Google Scholar]
- 17.Foner N. Immigrant women and work, then and now. In: Foner N, ed. In a New Land New York: New York University Press; 2005:89–105 [Google Scholar]
- 18.Greig FE, Koopman C. Multilevel analysis of women's empowerment and HIV prevention: quantitative survey results from a preliminary study in Botswana. AIDS Behav 2003;7:195–208 [DOI] [PubMed] [Google Scholar]
- 19.Foner N. Being Black in London and New York. In: Foner N, ed. In a New Land New York: New York University Press; 2005:109–130 [Google Scholar]
- 20.Kasinitz P, Battle J, Miyares I. Fade to black?: The children of West Indian immigrants in South Florida. In: Rumbaut R, Portes A, eds. Ethnicities: Coming of Age in Immigrant America New York, NY: Russell Sage Foundation; 2001:267–300 [Google Scholar]
- 21.Gil A, Wagner E, Vega W. Acculturation, familism, and alcohol use among Latino adolescent males: longitudinal relations. J Community Psychol 2000;28:443–458 [Google Scholar]
- 22.Krieger N. Discrimination and health. In: Berkman L, Kawachi I, eds. Social Epidemiology Oxford, England: Oxford University Press; 2000:36–75 [Google Scholar]
- 23. HIV/AIDS in New York City, 2004 . New York, NY: New York City Dept of Health and Mental Hygiene; 2006 [Google Scholar]
- 24.Wilson TE, Hogben M, Malka ES, et al. Reducing sexually transmitted infection rates by improving patient-based partner notification for treatment: a randomized controlled trial. Am J Public Health 98;8: In press [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Fishbein M, Triandis HC, Kanfer FH, Becker M, Middlestadt SE, Eichler A. Factors influencing behavior and behavior change. In: Baum A, Revenson TA, Singer JE, eds. Handbook of Health Psychology Mahwah, NJ: Lawrence Erlbaum Associates; 2001:1–7 [Google Scholar]
- 26. World Health Organization. AUDIT: Alcohol Use Disorder Identification Test. Available at: http://www.who.int/substance_abuse/activities/sbi/en/index.html. Accessed December 28, 2007.
- 27.Abraido-Lanza A, Dohrenwrend B, Ng-Mak D, Turner JB. The Latino mortality paradox: a test of the “salmon bias” and healthy migrant hypotheses. Am J Public Health 1999;89:1543–1548 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Abraido-Lanza AF, Chao MT, Florez KR. Do healthy behaviors decline with greater acculturation?: implications for the Latino mortality paradox. Soc Sci Med 2005;61:1243–1255 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Grant BF, Stinson FS, Hasin DS, Dawson DA, Chou SP, Anderson K. Immigration and lifetime prevalence of DSM-IV psychiatric disorders among Mexican Americans and non-Hispanic Whites in the United States. Arch Gen Psychiatry 2004;61:1226–1233 [DOI] [PubMed] [Google Scholar]
- 30.Hazuda H, Haffner SM, Stern MP, Kalichman SC. Effects of acculturation and socioeconomic status on obesity and diabetes in Mexican Americans. Am J Epidemiol 1988;128:1289–1301 [DOI] [PubMed] [Google Scholar]
- 31.Lucas JW, Barr-Anderson DJ, Kington RS. Health status, health insurance, and health care utilization patterns of immigrant Black men. Am J Public Health 2003;93:1740–1747 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Kenya SB, Divale MW, Allegrante JR, Fullilove RE. Effects of immigration on selected health risk behaviors of Black college students. J Am Coll Health 2003;52:114–120 [DOI] [PubMed] [Google Scholar]
- 33.Borrell LN, Castor D, Conway FP, Terry MB. Influence of nativity status on breast cancer risk among US Black women. J Urban Health 2006;83:211–220 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Hines AM, Caetano R. Alcohol and AIDS-related sexual behavior among Hispanics: acculturation and gender differences. AIDS Educ Prev 1998;10:533–547 [PubMed] [Google Scholar]
- 35.Newcomb MD, Romero GJ, Wayment HA, et al. Acculturation, sexual risk taking, and HIV health promotion among Latinas. J Counseling Psychol 1998;45:454–467 [Google Scholar]
- 36.Marin BV, Tschann JM, Gomez CA, Kegeles SM. Acculturation and gender differences in sexual attitudes and behaviors: Hispanic vs non-Hispanic White unmarried adults. Am J Public Health 1993;83:1759–1761 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Harawa N, Bingham T, Cochran S, Greenland S, Cunningham W. HIV prevalence among foreign- and US-born clients of public STD clinics. Am J Public Health 2002;92:1958–1963 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Sheeran P, Abraham C, Orbell S. Psychosocial correlates of heterosexual condom use: a meta-analysis. Psychol Bull 1999;125:90–132 [DOI] [PubMed] [Google Scholar]
- 39.Mathews C, Coetzee N, Zwarenstein M, et al. A systematic review of strategies for partner notification for sexually transmitted diseases, including HIV/AIDS. Int J STD AIDS 2002;13:285–300 [DOI] [PubMed] [Google Scholar]
- 40.Gonzalez Castro F. Is acculturation really detrimental to health? Am J Public Health 2007;97:1162 [Google Scholar]