International Journal of STD & AIDS >
Volume 19, Number 5 >
Pp. 309-315
Original research articles |
* Department of Obstetrics and Gynecology, The University of Texas Health Sciences Center San Antonio, San Antonio, TX, USA;
National Institutes of Health, National Institute of Allergy and Infectious Diseases, Bethesda, MD, USA
Correspondence to: Andrea Ries Thurman, Assistant Professor, Obstetrics and Gynecology, University of Texas Health Sciences Center, San Antonio 7703 Floyd Curl Drive, Mail Code 7836, San Antonio, TX 78229-3900, USA Email: thurmana{at}uthscsa.edu
 | Summary |
|---|
 Top Summary INTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
We used a cross-sectional analysis of 166 pregnant women with an STI, enrolled in a randomized controlled trial of behavioural intervention to prevent recurrent STIs. The primary outcome, PN, is notification of, or intent to notify male sexual partner(s) of STI exposure.
Pregnant women with one (n = 136) versus multiple (n = 30) partners reported PN for 88.2% and 54.5% of male partners, respectively (P < 0.001). Multivariate logistic regression demonstrated three variables that independently predicted PN: a steady relationship, with one male sexual partner and recent (<30 days) intercourse with the partner. Among the low income, pregnant MA and AA women, the three relationship variables predicted 81.6% of PN and correctly classified 78.5% of males notified and 65.7% of males not notified.
Key Words: partner notification • pregnancy • STD exposure
| INTRODUCTION |
|---|
|
TopSummaryINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
In this paper, we describe partner notification (PN) of STI exposure among a large cohort of pregnant, low income, mainly Mexican-American (MA) women. Much has been written about PN strategies among teens,5–9 and other populations at risk for STIs.10–24 However, we are not aware of any publications detailing PN specifically among pregnant, predominantly MA women.
The increasing demands placed on public health departments have limited comprehensive PN and treatment programmes.10,25 Increasingly, a woman's prenatal care provider must be skilled in counselling pregnant women with a STI on PN and partner treatment. PN and partner treatment strategies have become a major public health focus, with the Centers for Disease Control publishing guidelines on expedited partner therapy for STIs.26 PN involves partner elicitation by the index patient, contacting exposed individuals and providing testing and treatment.10,12,14,21 This process is complicated by several social, administrative, communication and psychological issues. Pregnancy further complicates the process, as it has been well described that pregnant women may be more vulnerable to relationship discord, abuse and domestic violence.27 Pregnant women may be reluctant to inform a male partner of an STI diagnosis, as she may depend on him emotionally, financially or she may be concerned about paternity.
The data for this study were obtained from a prospective, randomized controlled trial, which was designed to prevent recurrent STIs in low-income MA and African-American (AA) women.28 We obtained detailed, patient and partner information for this large cohort of pregnant women with STIs. In a prior analysis, we described PN in a large cohort of MA and AA women.29 Our goal in this study was to describe PN among pregnant women with one versus multiple sexual partners and to compare PN among pregnant and non-pregnant women. We aim to elucidate independent factors upon which clinicians can focus to identify pregnant women least likely to notify their partners, who may need additional assistance.
| METHODS |
|---|
|
TopSummaryINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
The primary outcome of the study, termed PN, was determined by asking each woman ‘Are you going to tell him that you have an STI so that he can get checked?’ for each male sexual partner. We considered: ‘Yes’, ‘He told me’, and ‘I told him already/he already knows’ to indicate ‘Yes PN’. We categorized: ‘Maybe’ and ‘No’ as ‘No PN’.
Each woman was asked to identify all male sexual partners in the last three months, the date of first and last intercourse with each partner, and whether any of the men were the father of the fetus. We obtained sociodemographic information (ethnicity, marital status, age, parity, employment, educational attainment, living situation and illegal drug, cigarette and/or alcohol consumption in the last three months), on the index patient. To determine if the woman perceived the relationship(s) as committed, we asked several questions, including: ‘Has your relationship been steady?’, ‘As far as you know, are you going to have sex with him again?’, ‘How important is it to keep this man?’ and ‘Did you want to get pregnant by this man?’
We asked each woman several questions regarding the pregnancy, including how she and her family felt about it when they first found out and how they felt currently, and how important it was to her to have a baby with her male sexual partner(s). We asked each woman several questions regarding stress and abortion considerations during this pregnancy. Finally, we inquired about plans for future childbearing and the importance of fertility to the patient and her male partner(s).
To determine whether a woman was currently experiencing emotional, physical or sexual abuse, we asked if any of the following situations were occurring right now and considered any ‘Yes’ answer to indicate a positive screen: ‘Has a man constantly criticized or put you down?’, ‘Tried to force you into doing something you didn't want to do by threatening you?’, ‘Acted with extreme jealousy, like he's your owner?’, ‘Used a knife, gun, or other weapon against you?’, ‘Forcefully held you down, punched, kicked or tried to choke you?’, ‘Made you have sex without protection against STIs, when you wanted protection?’, ‘Knowingly hurt you physically during sex?’, ‘Made you feel afraid to say no to sex?’, ‘Had sex with you when you didn't know what was happening or were out of control?’, ‘Made you have sex when you didn't want to?’.
To investigate ‘relationship issues’ we asked, ‘Respond ‘A lot’ ‘A little’ or ‘Not at all’ to the following questions: Besides sex, what do you get from your relationship with him: (1) emotional support, (2) housing/food, (3) other financial help, (4) feelings of warmth and closeness during sex, (5) feelings of warmth and closeness at other times, (6) drugs and/or alcohol and (7) physical protection’. For these questions, we categorized ‘A lot’ and ‘A little’ as ‘Yes’.
We divided the pregnant women into two groups: pregnant women with one versus multiple sexual partners in the last three months. Bivariate relationships between independent variables and PN were first explored using chi-square analysis. We then determined which variables were independently associated with PN using forward and backward-step multivariate logistic regression analysis. We then compared pregnant women (n = 166) with non-pregnant women (n = 606) enrolled in SAFE 228, stratified by number of partners and plans for future sexual contact with each man, to determine if PN was influenced by pregnancy status.
| RESULTS |
|---|
|
TopSummaryINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
|
Table 2 describes the sample and demonstrates that there was no difference among pregnant women with one versus multiple partners in baseline sociodemographic variables. Approximately one-third of pregnant women reported current emotional, physical and/or sexual abuse. Finally, the cohorts had similar responses regarding the importance of having children to them and their partner(s), thoughts of abortion during the current pregnancy, concern that a STI might harm the fetus and plans for more children. The pregnant women with multiple partners were initially more unhappy about being pregnant (P = 0.02); but when asked how they currently felt about the pregnancy, the two cohorts had similar feelings (P = 0.43).
|
|
|
We performed a forward step-wise, multivariate logistic regression analysis using the significant (P < 0.05) variables in Tables 3 and 4 including: number of partners, man lives with female, man is the father of the baby, relationship is steady, last intercourse was within 30 days, ongoing sexual activity is anticipated, woman believes it is important to ‘keep’ this man, he provides emotional support, physical protection, feelings of warmth and closeness during sex and/or at other times, female's education and employment status and female desires pregnancy with him. The resulting logistic regression analysis was confirmed using a backward step-wise method. Three variables were independently associated with PN: number of partners equals one (P = 0.003, adjusted odds ratio [OR] 2.44 95% CI 1.37, 4.35), relationship is steady (P = 0.007, adjusted OR 3.36 95% CI 1.39, 8.13) and time since last intercourse was 30 days or less (P = 0.001, adjusted OR 5.22 95% CI 1.95, 13.98). This model predicted 81.6% of PN and correctly classified 78.5% of males notified and 65.7% of males not notified.
Finally, a comparison was made between the 166 pregnant women and 606 non-pregnant women enrolled at the SAFE 2 intake visit. Table 5 demonstrates that pregnant and non-pregnant women responded similarly to questions regarding future sexual contact and PN. Most women, whether pregnant or not, would notify men with whom they anticipated future sexual encounters. The driver variable was number of male partners: women with one sexual partner, whether pregnant or not, had significantly higher rates of PN and were more likely to anticipate future sexual contact.
|
| DISCUSSION |
|---|
|
TopSummaryINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
The main issues which influenced PN in this sample centered on the relationship(s) with the male partner(s) and not the pregnancy. Even abortion considerations and other markers of stress and unhappiness about the pregnancy did not influence PN. PN was neither influenced by the woman's feelings regarding future fertility, the importance of childbearing, nor baseline sociodemographic characteristics. Women who perceived that they were in a committed relationship, namely, a ‘steady’ relationship, with one male partner, with whom they recently had intercourse, were more likely to report PN. Even variables that were initially predictive of PN, but did not remain independent in the final logistic regression, speak to relationship quality.
The United States Census Bureau reports that the fastest growing segment of the United States population is of Hispanic origin, mainly from Mexico. In 1990, 15.6% of all live births in the United States were to foreign-born women, and this percentage increased to 21.4% in 2000.30 In 1990, 75.7% of live-births to undocumented immigrants, the majority of whom were from Mexico, occurred in Texas, California, Florida, Illinois, New Jersey or New York.30 Our results contribute to the PN literature for pregnant, MA women with STIs in a large border state. Previous studies of PN and EPT have addressed predominately Caucasian11,13,17–19,22,31 or AA5–9,15,16,20,23,24 women. None of these studies reported on pregnant females as a separate cohort5–9,11,13,15–20,22–24 and some specifically excluded pregnant females.15 Our population reflects women most at risk for recurrent STI's: unmarried, minority women, under the age of 25.1
We divided the pregnant women into those with one versus multiple sexual partners because we hypothesized that these two groups would have different rates of PN. We confirmed this assumption, but found that the two cohorts had similar feelings regarding the current pregnancy, abortion considerations, fertility and the importance of bearing children. We found that being identified as the father was predictive of PN among women with multiple partners, but not among the women with one male partner, as most women with one partner identified this man as the father.
We found that pregnant women had similar concerns when considering PN as non-pregnant women. Our research is consistent with other studies that found women with one, steady partner were most likely to inform the man about the STI exposure.5,8,11,13,16–19,24,29,31 We, like others, found that index patients who anticipated ongoing sexual activity with exposed partners were more likely to disclose their infection.5,11,13,18,24,29,31 The recency of sexual intercourse has also been shown to positively influence PN.13,17,18,29
The Cochrane database reports that the domestic violence implications of PN are important research topics.21 Ongoing emotional, physical and/or sexual abuse was reported by approximately one-third of our cohort. It has been described that abuse escalates in pregnancy.27 None of the women in our sample reported that they would not notify a man because they feared physical violence, unlike our previous report.29 However, a small number of male partners (7%) were not notified because the pregnant women reported that they feared he would be angry. The perceived consequences of abuse and domestic violence after PN have been addressed in previous studies,5,6,9,13–15 but our study is unique to include a partner-specific abuse and violence screen. This screen has been described in this population, with regard to risk of pelvic inflammatory disease and other high-risk sexual behaviours associated with STI acquisition.32,33 We thought the question ‘Has any man made you have sex without protection against STIs, when you wanted protection?’ might be confounded by the fact that all women in the study were pregnant and recently diagnosed with a STI. However, only one of 49 women with positive abuse screens was included solely by this question.
The index woman's ethnicity, marital status, age and parity were not predictive of PN, as has been shown in other studies.8,9,11 We found that other variables that are particularly relevant to pregnant women including: illegal drug, alcohol and cigarette consumption, perceived financial independence, abortion considerations and current abuse were also not predictive of PN. Among pregnant women with multiple partners, women who were employed or had at least a high school education were significantly more likely to report PN, but this variable was not significant in the final logistic regression analysis, owing to the small cell size. Our PN rate among pregnant teens was similar to non-pregnant teens in other, smaller cohorts.5–9 Although other investigators found alcohol consumption influences PN,24 our pregnant population reported such low rates of alcohol, drug or tobacco use, that the PN analysis was limited by small cell size.
The questions that addressed relationship and communication issues predicted PN. However, number of partners, recent intercourse and a ‘steady’ relationship superseded the individual relationship and communication questions in the final model. Intercourse within 30 days was considered ‘recent’ because this is the recommended time to trace contacts after a diagnosis of Chlamydia trachomatis and Neisseria gonorrhoeae.1,14,21
PN is a difficult variable to validate. However, most male partners in our study fell into the categories of ‘I already told him/he already knows’ or ‘No’ I will not tell him. The cross-sectional nature of the study, although a limitation, reflects the reality a clinician faces when notifying a patient of STI and determining an effective method of PN.22 Another limitation of this study is that we did not contact any of the male partners to confirm our patient's answers. Of note, we have begun a third SAFE cohort that includes male partners.
Different PN strategies should be considered for each patient and determining an effective PN strategy is a complex process.12,21 Our data suggest that while a prenatal care provider may emphasize the pregnancy complications associated with STIs, what differentiates pregnant women who will notify their partners versus those who will not are relationship characteristics, specifically variables which indicate a committed partnership. We found three variables upon which a clinician can focus to determine which partners of pregnant women are most and least likely to be notified. Using these indicators, clinicians could provide additional counselling and assistance to pregnant women who are unlikely to notify their partners and help stop the cycle of re-infection.
| ACKNOWLEDGEMENTS |
|---|
|
TopSummaryINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
| REFERENCES |
|---|
|
TopSummaryINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||
