THE SPREAD OF SEXUALLY transmitted diseases (STDs) and their sequelae can be substantially reduced by early diagnosis and prompt interventions. Greater incidence and the chronic nature of STD sequelae in women make care‐seeking behavior of women an important public health priority. Among men and women, barriers to seeking care for STDs may include cost, embarrassment, secrecy, and geographic distance to a service provider. Additionally, STDs among women are more likely to be asymptomatic than STDs among men, thus creating an added barrier to seeking care for STDs among women.1 Women may also delay seeking care for a suspected STD longer than men.2
Recent studies also suggest that idealized images of relationships with men may drive women to believe their steady partner is not infected by an STD or human immunodeficiency virus (HIV) and is not at‐risk of becoming infected.3–5 Given this possibility, a woman who has sex with only her steady partner may be unlikely to seek care for a suspected STD or HIV infection based on a denial that she is at‐risk of infection from her partner. Belief in the “protective value” of a steady partner may be a product of American ideals dictating that sex is a display of trust and connectedness to the partner who reciprocates through sexual fidelity. These ideals are often emphasized in the acculturation of women and therefore may be a phenomenon common to U.S. women regardless of demographic differences.6 Further, women may be more likely than men to learn gender roles equating self‐worth with an ability to select and attract a valued sex partner: someone who is healthy (not infected with an STD or HIV) and monogamous.4
The type of clinic providing diagnostic and intervention services may also be a barrier for women seeking care for suspected STD or HIV infection. Data from the National Health and Social Life Survey indicated that STD clinics were rarely named as locations for STD treatment.7 Because women of reproductive age are particularly likely to interface with providers of gynecologic services (i.e., private practitioners and community health clinics) and family planning services, it is likely that women may prefer combining their STD/HIV care needs with other health care needs. Yet, embarrassment and perceived unprofessionalism have been identified as common barriers to seeing a general practitioner about a suspected STD.8 Convenience, low cost, and a high regard for confidentiality have been identified as common reasons for choosing to seek care at an STD clinic.9 Preference for type of clinic may be a function of income, with low‐income women facing unique obstacles to health care: lack of transportation, lack of health insurance, and lack of child care.
The purpose of this study was to identify barriers to seeing a doctor about a suspected STD or HIV infection among a community sample of low‐income women reporting a history of at least one STD. In particular, we sought to determine if women did not seek medical care because they denied the possibility of an STD or HIV infection based on a belief in the protective value of a steady male partner. We hypothesized that this belief would be common among women regardless of urbanicity, race, ethnicity, income level, and education. Additionally, we identified preferred types of clinics for STD/HIV testing among low‐income women and hypothesized that women would be unlikely to prefer STD clinics as testing sites, regardless of urbanicity, race, ethnicity, income level, and education.
Data Collection and Analysis
A paper and pencil survey of women attending Women, Infants, and Children (WIC) clinics in Missouri was conducted jointly by the Bureau of STD/HIV Prevention in the Missouri Department of Health and the Rural Center for AIDS/STD Prevention at Indiana University. Twenty‐nine WIC clinics located in 21 counties were selected by stratified random sampling. We stratified by size of county, selecting 15 rural counties, 4 suburban counties, and the only 2 urban counties of Missouri. Human subjects' approval was granted by Missouri Department of Health and Indiana University.
From February through April of 1998, women 18 years of age or older attending WIC clinics were solicited by clinic staff to participate in the survey. WIC guidelines stipulated that attendees must earn an annual income not exceeding 185% of the poverty definitions established by the U.S. government. Women agreeing to participate were provided with a preaddressed, postage‐paid envelope for return of a brief questionnaire. Spanish translations of the questionnaire were available. Questionnaires were coded for size of county where the WIC clinic was located. Counties of 50,000 people or less were classified as rural, with counties of 500,000 or more being classified as metropolitan (urban), and the remainder being classified as small metropolitan areas (suburban). Response rates were 80% for rural counties, 49% for urban counties, and 41% for suburban counties. The overall response rate was 58%, representing 2,256 women.
Computer software was used to generate frequency distributions of self‐reported STDs and HIV infection. Subsequent analyses were based on a subsample of respondents indicating a history of at least one STD or indicating an HIV‐positive serostatus. Women were asked “In the past, if you did not see a doctor about an STD or HIV, why not?” and provided with a list of nine possible barriers to seeing a doctor. Women were instructed to indicate multiple barriers if applicable. Women were also asked, “If you ever worried that you might have an STD or HIV where would you most likely be tested?” and provided with a list of four alternatives plus an “I don't know” alternative, an alternative reading, “I would not go anywhere,” and an alternative indicating they have never worried about STD/HIV. Women were instructed to indicate one answer only. Focus groups of low‐income Missouri women judged the questions as being easy to read and understand.
Forward stepwise logistic regression was used to determine the contributions of urbanicity, race, ethnicity, income, and education to reporting that medical care was not sought because “I only had sex with my steady” and to preference for STD/HIV testing location. Alpha was set at 0.05.
Description of the Sample
The majority of the respondents in the community sample were rural (56.3%), with the remainder being composed of urban (18.5%) and suburban (25.2%) women. Median age was 25 years. The sample was mostly white (75.3%), with 21.3% black participants, and the remainder being composed of Asian/Pacific Islanders (0.4%), American Indians (1.2%), and persons of mixed race (1.8%). Hispanic ethnicity was reported by 5.4% of the respondents. Median monthly income was $601 to $1,200. Median level of education was completion of high school. Less than 3% of the respondents reported sex with more than one male partner in the past 2 months, with about 88% reporting sex with one male partner in the past 2 months.
More than one fifth (21.5%) of the community sample reported a past history of at least one STD (n = 484), and seven women reported being HIV‐positive. These 491 women comprised the subsample for these analyses. Of these women, the majority (77.6%) reported having only one STD, 18.3% reported having two STDs, and 3.5% reported having three or more STDs. Chlamydia was the most commonly reported STD (61.7%), followed by gonorrhea (25.3%), human papillomavirus, (20.4%), herpes (11.6%), syphilis (6.1%), and HIV (1.4%). Table 1 displays the percentages of STDs and HIV among women from the community sample and women from the subsample.
The subsample was predominately white (55%) and black (40.5%), with 1% being Asian/Pacific Islanders and 3% being of mixed race. The majority of the subsample was non‐Hispanic, with only 4.7% indicating Hispanic ethnicity. Median age was 24 years. Five income intervals ranged from $0 to $600 to $3,000 or more per month, with the median income interval being $601 to $1,200 per month. Median level of education was completion of high school. Women from rural counties comprised 41.4% of the subsample, with women from small metropolitan areas comprising 22.8%, and women from metropolitan areas comprising 35.5% of the subsample. Chi‐square tests indicated that black women were overrepresented in the subsample (40.5% versus 21.3%; χ2 = 80.5, df = 4, P < 0.001) and urban women were overrepresented in the subsample (35.5% versus 18.5%; χ2 = 66.6, df = 2, P < 0.001).
Regarding possible barriers to seeing a doctor, the majority (58.7%) of the subsample indicated that none of the nine possible barriers kept them from seeking treatment. Twenty percent of the subsample did not respond to this item. Table 2 displays frequencies of responses to the nine barriers for the remaining 21.3% of the sample (n = 105). Among women indicating at least one barrier to seeing a doctor about STD or HIV, the most frequent response was “I only had sex with my steady” (36.2%), followed by “I did not have any symptoms” (33.3%). Cost (25.7%) and embarrassment (22.8%) were also selected by a relatively large portion of the women who indicated at least one barrier to seeking care from a doctor. Other barriers were selected less frequently. Logistic regression analysis indicated that urbanicity, race, ethnicity, income, and education did not predict belief in the protective value of a steady male partner.
Table 3 shows preference for STD/HIV testing location among women reporting a history of STD. Approximately 10% of the subsample indicated they never worried about STD/HIV, and nearly 5% indicated they would not know where to go for testing. Only one woman said she would not go anywhere for testing. Ten women did not answer this question. Of 411 women indicating a preference for one of four testing locations, the majority (63.8%) indicated a preference for being tested by their doctor. Fewer women reported a preference for being tested in a community health center (14.8%) or a family‐planning clinic (16.8%). A relatively small portion (4.6%) of the sample indicated a preference for testing at an STD clinic.
Logistic regression analyses indicated that preference for testing location was not predicted by urbanicity, income, or ethnicity. Education predicted preference for “my doctor” (OR = 1.24; 95% CI = 1.04‐1.47) and black race predicted preference for STD clinic (OR = 5.34, 95% CI = 1.10‐25.94).
Among women reporting barriers to seeking care for a suspected STD or HIV infection, the most commonly reported barrier was “I only had sex with my steady.” With the exception of being asymptomatic, no other barrier was indicated nearly as often by women in the subsample. The finding implies that women may have an inherent trust in their male partners' sexual health and fidelity and subsequently attribute possible symptoms of STD to some other cause, e.g., candidiasis, menstrual pain, or bladder infections. This finding was not confounded by the possibility that women feared their partner might discover the diagnosis (or the clinic visit) because a separate alternative was provided for women in regards to this fear. Believing in a protective value of a steady partner was not predicted by urbanicity, race, ethnicity, income, or education. Thus, the belief appears to cut across a broad spectrum of U.S. women reporting an income of less than 185% of poverty.
In addition, women who have experienced at least one STD or have been diagnosed with HIV report an overwhelming preference for being tested for STD/HIV by their own doctor. This finding was expected in that women are likely to have an established relationship with a health care provider whom they see for other health care needs. An established relationship with these providers probably facilitates women's ability to disclose their symptoms or fears about STD and/or HIV infection. Finding that embarrassment was a common barrier to seeking care for a suspected STD or HIV infection suggests preference for STD clinics may have been low due to a stigma attached to attendance at these clinics as opposed to clinics offering a range of health care services. Women in our subsample were generally similar with regard to preference for type of clinic providing STD/HIV services, with differences in urbanicity, levels of income, and ethnicity not associated with clinic preferences.
Other findings were also important. Our study showed that most women did not perceive any of the nine barriers provided as keeping them from seeking care for an STD or HIV infection. This finding is instructive in that a majority of women with a history of at least one STD or HIV may perceive STD/HIV care services as being accessible, implying success of service providers in this regard. A comparison of the demographics described for the community sample and the subsample was also instructive. A greater percentage of urban women and a lower percentage of rural women were in the subsample reporting a past STD/HIV than in the community sample. Also, a greater percentage of black women and a lower percentage of white women were in the STD/HIV subsample than the percentage in the community sample. Thus, black women from urban areas are overrepresented relative to self‐reported history of STD and HIV infection (of the seven women who reported being HIV‐positive, four were black, two were of mixed race, and five were from urban counties).
Limitations and Implications
Study results are limited by the validity of self‐reported history of STD and HIV infection as well as the validity of the questions assessing barriers to seeing a doctor and preference for STD/HIV testing location. Findings must be interpreted in the context of women perceiving barriers to seeking care about a suspected STD or HIV infection as opposed to a previously diagnosed infection. Also, women attending WIC clinics were asked to participate in a survey about women's health practices, thereby creating a potential for self‐selection bias. Low response rates from women in urban and suburban counties also created a potential for self‐selection bias. Finally, the logistic regression analyses were based on low sample sizes and may therefore have yielded unreliable measures of association.
Overall, the findings supported our hypotheses. However, with regard to a common belief in the protective value of a steady partner, an alternative explanation exists. These women may have engaged in defensive denial relative to accepting the possibility that a steady partner infected them with an STD or HIV. In other words, some women may have suspected that their partner was not monogamous, but they did not think about this possibility at a salient level. However, we provided a response alternative for women who may have been experiencing defensive denial, i.e., “I did not want to deal with it.” Approximately 11% selected this response alternative, showing a separation of defensive denial from belief in the protective value of a steady partner.
Within these limitations, the data indicate a need to design prevention messages that address common barriers to seeking care for a suspected STD or HIV infection. For example, messages emphasizing to women the value of seeking care for a suspected STD or HIV infection even if they are in a steady, seemingly monogamous, relationship may be useful. These messages can be constructed to include information about the possibility of monogamous partners importing an undiagnosed STD or HIV infection from a pervious relationship. Further, the data support offering affordable STD/HIV‐related services in locations already providing other health services for women.