Chlamydia trachomatis and Neisseria gonorrhoeae are the most common reportable sexually transmitted infections (STI) in the United States.1 Men and women of African descent bear a disproportionately higher burden of STIs in the United States, with disparities being greatest for bacterial STIs such as gonorrhea and chlamydia.2 With regard to Healthy People 2010 objectives, the 5 greatest disparities between men and women identified as non-Hispanic black as compared with white all relate to human immunodeficiency virus (HIV) and STI infections.3
Previous research has suggested that racial and ethnic disparities in HIV/STI may be due in part to factors such as poverty and income-inequality.4,5 For instance, women at greatest risk for acquiring or transmitting STIs are disproportionately living at or below the poverty level.6 In particular, black women are more likely to live in poverty where there is a higher prevalence of HIV infection and injection drug use.7 Moreover, violence, racism, lack of economic opportunity, and incarceration may serve to limit potential male partners, which could result in an increase in the value of having a partner and contribute to male dominance in intimate relationships.8 As described in prior studies, relationship power plays a significant role in safer sex decision making and HIV/STI risk.9 To date, efforts to promote safer sexual practices may be limited to the extent to which they assume a degree of individual control that does not exist for many women.10 Conversely, when efforts to increase condom use address social, economic, and cultural realities of women's lives and the context of their relationships, they are likely to be more effective.8
In a previous publication, we reported on a behavioral intervention designed to reduce sexual risk behavior and increase STI partner notification among men and women diagnosed with gonorrhea or chlamydial infection in 2 urban clinics serving primarily low income, minority patients.11 As part of this intervention, we developed components designed to improve attitudes, skills, and perceived norms supporting STI partner notification and sexual risk reduction. Specifically, some of the content was informed by formative work that revealed barriers to behavior change including fears of loss of sexual partners, perceived concerns of stigmatization, and worries about arguments that might follow behaviors such as STI partner notification and condom negotiation with partners. We attempted to address these barriers through a number of activities that sought to reduce the likelihood that these consequences would occur, including improved negotiation skills surrounding condom use, approaches to preventing or diffusing arguments with sex partners, and methods to reduce blame and stigmatization. Evaluation of the resulting program demonstrated statistically significant increases in self-reported partner notification, decreases in sexual risk behavior, and a lower risk for infection with chlamydia and/or gonorrhea at a 6-month follow-up.
As part of the evaluation for that program, we assessed participants' perceptions that they were able to meet basic needs. Although a measure of perceived basic need is a subjective estimate, such a measure can be viewed as more likely to inform decision making regarding sexual behavior than would more distal indices such as comparisons with living income standards or poverty. The measurement of perceived basic needs used in the current study includes assessment of resources by which an individual can fulfill fundamental needs such as food, shelter, and clothing. Individuals, particularly women, who have higher levels of unmet basic needs, may be less likely overall to reduce HIV/STI transmission risk following an STI diagnosis. This could either be because they are concerned that negotiation around safer sex may jeopardize a relationship that contributes to fulfilling basic financial needs, or it could also reflect the relatively low priority that a treatable STI might present in the context of these higher needs and a life that may be characterized by increased disorder. There is a body of literature that indicates that individuals with competing needs and high levels of chaos in their lives are less likely to seek medical care and adhere to medical appointments.12–14 However, the effect of life chaos has not been investigated in terms of its impact on sexual risk behavior. A program that seeks to reduce risk of loss of partnerships or conflict in relationships might attenuate the relationship between basic needs and sexual risk reduction. The purpose of the present analysis is thus to explore the relationship between perceived basic needs and sexual risk behavior following an STI diagnosis, and to examine whether receipt of an HIV/STI risk reduction program influences the relationships between perceived basic needs and risk behavior.
A cross-sectional assessment of data on perceived financial resources were collected and analyzed as part of a randomized controlled trial of an intervention aimed at promoting timely STI partner notification and reducing sexual risk behavior among patients presenting for care at 1 of 2 STI treatment centers in Brooklyn, NY. Eligible participants were 18 years or older, reported being sexually active in the past 2 months, could complete a self-report interview in English or Spanish, and had been diagnosed by a health care provider with either gonorrhea or chlamydial infection. In addition, microbiologic confirmation of C. trachomatis or N. gonorrhoeae was required, based on the diagnostic methods utilized at each clinic.
As part of study participation, all participants completed a standardized self-report interview at baseline and 6 months, administered by centrally trained research assistants. Cross-sectional data on perceived financial resources were collected from 600 participants enrolled between January 2002 and October 2004, and 533 (89%) completed a 6-month follow-up assessment. Of the 533 who completed both study visits, we removed 5 participants from analysis with missing data on either sexual risk behavior or financial needs. All program activities were approved by institutional review boards at the participating sites and at the Centers for Disease Control and Prevention. On average, participants received $70 for completing all elements of the study.
Perceived financial needs were assessed using the 7-item Basic Needs subscale of the Family Resource Scale-Revised (FRS-R).15 The FRS-R is based on the Family Resource Scale,16 but was revised and psychometrically validated using a large sample of low-income, ethnic minority adults. The items in the Basic Needs subscale reflect the participant's perception of the degree to which s/he had enough food for 2 meals a day for the family, a house or apartment for the family to live in, enough clothing for the family, enough heat for the house or apartment, indoor plumbing and water, enough furniture for the house or apartment, and access to a telephone. For each item, the participant responded on a 5-point scale as to whether s/he consistently had these items “never enough,” “seldom enough,” “sometimes enough,” “usually enough,” or “almost always enough.” The FRS-R was implemented at the 6-month follow-up given that we were interested in how perceptions of needs might be linked to sexual behavior assessed during the follow-up assessment. Cronbach's α for the subscale in our population was 0.86. Women who reported that they “usually” or “almost always” had enough on each of the 7 items were categorized as having “met needs,” and those who reported at least once that they “never,” “seldom,” or “sometimes” had enough of at least one item were categorized as having “unmet needs.” In our study, the scale was implemented at the 6-month follow-up given that we were interested in how current perceptions of needs might be linked to sexual behavior assessed during the follow-up assessment.
Sexual behavior and sexual risk behavior were also assessed at the 6-month interview. Participants reported on their sexual behavior and condom use consistency during anal and vaginal intercourse over the last 90 days. In the interview, questions were asked separately for main (regular), casual, and one-time partners. Responses across all 3 types of partners were collapsed to form an index reflecting any episode of unprotected anal or vaginal intercourse (UAVI) over the last 90 days. The variable was dichotomized such that 1 group contained participants who reported no sexual activity or reported consistent condom use during vaginal and anal sex, and the other group contained participants who had at least 1 episode of UAVI.
Several other variables from the baseline interview that have been found to associate with sexual risk behavior were also included in this analysis. Demographic data included participant sex, age, whether the participant was born in the United States, racial and ethnic self-identification, and marital status. Number of sex partners at baseline was categorized into 0, 1, or >2 partners. Household income was defined as the total annual amount of money that the participant and his/her family had to live on (before taxes) from all possible sources. This variable was dichotomized into those whose annual household income was greater than or less than or equal to $12,000. Education was dichotomized into those that received a high school diploma or GED and those who did not. Employment at baseline was dichotomized as those who were employed either part or full time and those who were not.
Our analysis used a cross-sectional design—the main predictor and outcomes variables in it were assessed at one time point (i.e., the 6-month follow-up of the intervention). Bivariate relationships between posited predictors of perceived basic needs and UAVI were assessed using chi-square analysis. A multivariable logistic regression analysis was performed to examine the relationships between UAVI and perceived basic needs (unmet needs vs. met needs), controlling for age, sex, site of recruitment, intervention group membership, country of origin, education level, and employment status. We applied a correction method to the adjusted odds ratios (AOR) to account for the relatively high incidence of UAVI in the sample.17 To examine moderator effects, we created interaction terms by dummy-coding variables such as intervention group and included the product terms in the regression model. We conducted a stratified analysis to examine how perceived basic needs were related to UAVI in both the group that received the intervention and in the group that did not, and specifically to examine relationships between sex and perceived needs in these 2 groups by further stratifying by sex among individuals who did not receive the intervention. These analyses involved controlling for site of recruitment, age, and whether the participant was born in the United States. All analyses were performed using SPSS Version 17.0, with statistical significance set at a 2-sided α level of 0.05.
Baseline Participant Characteristics
The mean age of participants at study entry was 25 years (standard deviation = 6.7); 40.5% were women. Participants self-identified primarily as either Caribbean (52%) or black (40%); 10.5% reported being Hispanic. Nearly half (45%) of participants were born outside of the United States, 64% had completed high school or an equivalent, 68% had received income from a job in the past month, and 53% reported an annual household income of at least $12,000 per year. Twenty-eight percent were married or living with a partner and 54% reported more than one partner in the past 3 months at baseline. Fifty-one percent of participants were randomized to the group that received the intervention. Participants randomized at baseline to the group that did not receive the intervention (N = 296) did not differ from those assigned to the group that did (N = 304) as a function of age, sex, racial or ethnic identification, birthplace, education, employment, income level, partnership status, or number of sexual partnerships (all P > 0.05). Of the 528 cases analyzed in our analysis, 249 received positive test results for chlamydial infection at baseline, 208 for gonorrheal infection, and 71 for both infections. All baseline chlamydial infections were confirmed through NAAT testing. Of the 208 for gonorrhea only, 137 were based on NAAT; the remaining 71 utilized other methods (Gram stain, culture).
Perceived Basic Needs
Nearly half of respondents believed that they had enough or almost always enough food for 2 meals a day for the family (63%), a house or apartment for the family to live in (64%), enough clothing for the family (60%), enough heat for the house or apartment (63%), indoor plumbing and water (66%), enough furniture for the house or apartment (54%), and access to a telephone (53%). Overall, 57% percent of respondents reported that their financial needs were met either “almost always” or “usually” across all 7 items, and were categorized as having their needs met; the remaining 43% were categorized as having unmet needs. As expected, perceived basic needs were associated with both education level and current annual household income (Table 1). In addition, a higher proportion of foreign-born respondents were categorized as having unmet needs, compared with those born in the United States (P = 0.02). Of note, 51% of the group of participants that received the intervention measured at follow-up fell into the unmet needs category, as compared to 35% of participants who did not receive the intervention (P < 0.001).
Perceived Basic Needs, Intervention Group, and Sexual Risk Behavior
Of the 528 respondents, 57% reported UAVI at the 6-month follow-up interview (301/528). Although there were no differences in UAVI as a function of assigned intervention group at baseline (P = 0.65), there were differences at follow-up such that those who received the intervention were less likely to report UAVI (52%) as compared with those who did not (62%; P = 0.02), as reported in a prior publication.11 Collapsing across intervention conditions, 6-month reports of UAVI were associated with basic needs; those who were categorized as having unmet needs were more likely to report UAVI (62%) versus those who had met needs (53%; P = 0.03). As shown in Table 2, the relationship between basic needs and UAVI remained statistically significant (AOR = 1.28; 95% confidence interval = 1.04–1.53), after controlling for other covariates in the model. In this model, receipt of the intervention was also associated at statistically significant levels with UAVI after controlling for age, sex, site of recruitment, unmet needs, and country of origin. A second model that included income and education level in addition to these covariates did not influence the significance or strength of these relationships. A test of the 3-way interaction between intervention group, sex, and basic needs on UAVI was not statistically significant (P = 0.23).
We conducted a stratified analysis to examine how perceived basic needs were related to UAVI among those who received the intervention and those who did not, and specifically to examine relationships between sex and perceived needs in these 2 groups, controlling for site of recruitment, age, and whether the participant was born in the United States. In the group that received the intervention, there was no effect of any of our predictor variables on UAVI, nor was there a significant interaction between sex and perceived need (all P > 0.05). However, in the group that did not receive the intervention, the interaction between sex and basic needs was statistically significant (P = 0.049). An additional stratification by sex among those that did not receive the intervention was conducted. The subgroup analysis involving women only showed an effect of perceived needs on risk of UAVI (AOR = 1.18; 95% confidence interval = 1.07–1.24) such that women who did not receive the intervention who were categorized as having unmet needs were more likely to report any UAVI (78%) as compared with women who have met needs (54%) after controlling for age, sex, site of recruitment, and country of origin. However, in a parallel analysis of men who did not receive the intervention, those with unmet needs were no more likely to report UAVI (63%) as compared to those with needs met (60%; P = 0.70).
Researchers and advocates in the areas of women's health and HIV/STI prevention have long noted the importance of examining and addressing how poverty and socioeconomic disparities may be a key limiting factor in the success of traditional intervention programs that rely solely on modification of intrapersonal factors. In this analysis, we sought to explore the relationship between perceived basic needs and sexual risk behavior among urban, minority men and women recently diagnosed with gonorrhea and chlamydial infections.
Our analysis suggests that in this population of STI clinic patients, there is a high level of unmet basic need, with over 40% of participants falling into this category. As expected, reports of unmet basic need at the follow-up interview were associated with baseline socioeconomic indices such as education and household income. Reports of unmet needs in this population reflect the socioeconomic realities of these population subgroups; women were more likely to report unmet needs as compared to men, and foreign born participants, who were mostly from the Anglophone Caribbean, were more likely to report unmet needs as compared with those born in the United States.18,19
Those who were categorized as having unmet needs in our analysis were more likely to report engaging in unprotected vaginal or anal sex at the follow-up interview occurring 6 months after a baseline diagnosis with gonorrhea or chlamydial infection, controlling for intervention group membership and other covariates. However, further analysis revealed that this effect was driven by differential rates of unprotected sex among women in the group that did not receive the intervention. No such difference was detected among women in the group that received the intervention. One plausible explanation is that the intervention program may have had some success in addressing women's concerns about the potential negative outcomes associated with suggesting changes in condom use patterns with sexual partners, thereby mitigating the influence of power or income differentials on sexual behaviors.
This interpretation, however, needs to be weighed against several possibilities. Unmet needs were higher among those women who received the intervention versus those who did not. One possible explanation for this is differential attrition; those with higher perceived needs who received the intervention may have been deterred by the behavioral strategies proposed and therefore chose to drop-out of the evaluation. This alternative explanation cannot be ruled out, given that perceived needs were not assessed at baseline. Another possible explanation is that some aspect of the intervention impacted women's perception of their own financial need; this related to our finding that, although perceived needs differed by intervention group, there were no differences at the same time point as a function of intervention group membership in terms of income (P = 0.11) or education level (P = 0.31). These and other possible explanations require further investigation.
Our design had additional limitations that affect interpretation of findings. First, unprotected sex is only a risk factor for HIV/STI acquisition if the person's partner is infectious. However, it is probably a reasonable measure of acquisition risk in this study because the participants were all recently infected with an STI. Second, there are inherent problems with using a self-report measure of sexual risk behavior as it is potentially prone to social desirability bias and recall bias. Third, we did not measure basic needs at baseline, which did not allow us to consider potential causal relationships among the variables. Fourth, we conducted a number of post hoc analyses that may have contributed to an overestimation of significant results. Finally, given that 30% of eligible respondents declined to participate in the study, the results of this study are not necessarily generalizable to the population of GC/CT infected STI patients.
Our findings present preliminary evidence that HIV/STI intervention components that seek to directly deal with issues of reduction in partner conflict might be beneficial to women with high perceived unmet basic needs, and for whom a potential dissolution of a relationship may represent a further loss in ability to meet basic needs. Future research to examine these issues would benefit from more direct assessment of resource and financial reliance on sexual partners as a factor that may influence behaviors after exposure to HIV/STI prevention programs.
1. Centers for Disease Control and Prevention. Sexually Transmitted Disease Surveillance, 2008. Atlanta, GA: Department of Health and Human Services, Centers for Disease Control and Prevention, 2009.
2. Newman LM, Berman SM. Epidemiology of STD disparities in African American communities. Sex Transm Dis 2008; 35:(suppl 12):S4–S12.
3. Keppel KG. Ten largest racial and ethnic health disparities in the United States based on healthy people 2010 objectives. Am J Epidemiol 2007; 166:97–103.
4. Adimora AA, SchoenbachVJ, Martinson FE, et al. Heterosexually transmitted HIV infection among African Americans in North Carolina. J Acquir Immune Defic Syndr 2006; 41:616–623.
5. Aral SO, Adimora AA, Fenton KA. Understanding and responding to disparities in HIV and other sexually transmitted infections in African Americans. Lancet 2008; 372:337–340.
6. Stratford D, Mizuno Y, Williams K, et al. Addressing poverty as risk for disease: Recommendations from CDC's consultation on microenterprise as HIV prevention. Public Health Rep 2008; 123:9–20.
7. El-Bassel N, Caldeira NA, Ruglass LM, et al. Addressing the unique needs of African American women in HIV prevention. Am J Public Health 2009; 99:996–1001.
8. St. Lawrence JS, Eldridge GD, Reitman D, et al. Factors influencing condom use among African American women: Implications for risk reduction interventions. Am J Community Psychol 1998; 26:7–28.
9. Pulerwitz J, Amaro H, De Jong W, et al. Relationship power, condom use and HIV risk among women in the USA. AIDS Care 2002; 14:789–800.
10. Krishnan S, Dunbar MS, Minnis AM, et al. Poverty, gender inequities, and women's risk of human immunodeficiency virus/AIDS. Ann NY Acad Sci 2008; 1136:101–110.
11. Wilson TE, Hogben M, Malka ES, et al. A randomized controlled trial for reducing risks for sexually transmitted infections through enhanced patient-based partner notification. Am J Public Health 2009; 99(suppl 1):S104–S110.
12. Wong MD, Sarkisian CA, Davis C, et al. The association between life chaos, health care use, and health status among HIV-infected persons. J Gen Int Med 2007; 22:1286–1291.
13. Penniman TV, Taylor SL, Bird CE, et al. The associations of gender, sexual identity and competing needs with healthcare utilization among people with HIV/AIDS. J Natl Med Assoc 2007; 99:419–427.
14. Cunningham WE, Andersen RM, Katz MH, et al. The impact of competing subsistence needs and barriers on access to medical care for persons with human immunodeficiency virus receiving care in the United States. Med Care 1999; 37:1270–1281.
15. Van Horn M, Bellis J, Snyder SW. Family resource scale-revised: Psychometrics and validation of a measure of family resources in a sample of low-income families. J Psychoeduc Assess 2001; 19:54–68.
16. Dunst CJ, Leet HE. Measuring the adequacy of resources in households with young children. Child Care Health Dev 1987; 13:111–125.
17. Zhang J, Yu KF. What's the relative risk? A method of correcting the odds ratio in cohort studies of common outcomes. JAMA 1998; 280:1690–1691.
18. Cobb-Clark DA, Hildebrand V. The wealth and asset holdings of US-born and foreign-born households: Evidence from SIPP data. Review of Income and Wealth 2003; 52:17–42.
19. Semega J. Men's and women's earnings by state: 2008. American Community Survey. Washington, DC: US Census Bureau, 2009.