Abstract
Curable sexually transmitted infections (STIs) including Chlamydia trachomatis (CT), Neisseria gonorrhoeae (NG), and Trichomonas vaginalis (TV) are associated with adverse pregnancy outcomes. Partner notification is an important component of STI control as it has been shown to prevent re-infection and reduce infectious burden. Between October 2017 and February 2019, we conducted a cohort study of women attending antenatal care in Cape Town, South Africa. Self-collected vulvovaginal swabs were tested for CT, NG, and TV using Xpert® assays at first antenatal visit, during the third trimester, and postpartum. At the visit following a positive diagnosis, women were asked if they notified their partner and if their partner was treated. Among 242 participants, 97% reported being willing to notify partners if they tested positive and 78% thought their partner would be willing to treat the STI. Of the 73 women who were diagnosed with one or more STIs and reported having a sex partner, 93% reported notifying their partner and 63% reported their partner was treated. Younger maternal age was associated with partner notification and treatment (OR = 3.82; 95%CI = 1.34–10.90). Acceptability of partner notification was high in pregnant women, but partner treatment was low. Future interventions to improve partner notification and treatment are needed.
Introduction
Chlamydia trachomatis (CT), Neisseria gonorrhoeae (NG), and Trichomonas vaginalis (TV) are among the most common sexually transmitted infections (STIs) globally. According to the World Health Organization, 370 million new cases of these three curable STIs are estimated to occur each year among adults aged between 15 and 49 years. 1 In pregnancy, untreated STIs are associated with adverse outcomes including stillbirth, preterm labor and delivery, and low birth weight.2–4 Some curable STIs can also be vertically transmitted to the neonate during birth, causing conjunctivitis and pneumonia. 5 In addition, STIs have been shown to increase the risk of Human Immunodeficiency Virus (HIV) acquisition and mother-to-child transmission, and the risk is highest in women with multiple STIs. 6 , 7
Partner notification is an important component of STI management and control because successful notification and treatment can prevent reinfection of the index patient (patient diagnosed with a STI) and reduce the burden of these infections. 5 Due to cost and time associated with expedited partner therapy, many low- and middle-income countries use a patient-led partner notification strategy where the index patient informs their own partner and refers them to treatment. 5 , 8 However, in a recent systematic review of studies conducted in sub-Saharan Africa, the proportion of index STI patients who notified their partner using this strategy was only 53% (range 23–95%). 9 Of those who successfully notified their partner, 25% (range 0–77%) had partners that sought treatment. 9 In another study at a community-based clinic in South Africa, participants reported notifying only 64% of partners of their STI diagnosis. 10 Among South Africans, fears of intimate partner violence and relationship dissolution have been identified as barriers to partner notification following diagnosis with a STI. 11 , 12 Others have been conducting research on how best to optimize STI management, including partner notification, in South Africa since the late 1990s.13–16
Although the prevalence and incidence of curable STIs in pregnancy are high in sub-Saharan Africa, few studies in the region have examined the acceptability of partner notification in pregnant women.17–20 One study of pregnant women in Botswana found that while 98% of participants at baseline said they would be willing to notify their partner if they tested positive for CT, NG, or TV, only 84% reported actually doing so and 63% reported that their partner received STI treatment. 20 Our study sought to better understand partner notification in pregnant women following STI diagnosis to inform interventions and policy makers about how best to provide STI diagnostics and treatment in pregnancy. We evaluated the acceptability of partner notification for STIs among pregnant women and examined factors associated with both partner notification and successful partner notification and treatment in antenatal care in Cape Town, South Africa.
Methods
Between October 2017 and February 2019, we conducted a cohort study of pregnant women attending antenatal care at a public sector clinic in Cape Town, South Africa. Our study was conducted in a primary antenatal care clinic at a large community-based public sector facility in Cape Town, South Africa, which serves a predominantly isiXhosa-speaking African population of approximately 350,000 people with an antenatal HIV prevalence of 30%. 21 Data collection, specimen collection and testing, and treatment have been previously described. 17 Briefly, women ≥18 years and currently pregnant (<34-weeks) were eligible to participate in the study. Women participated in three visits throughout their pregnancy: at first antenatal visit, during the third trimester, and post-partum. At each study visit, trained staff administered surveys to collect self-reported information on participant demographics, sexual behavior during pregnancy, partner history, history of intimate partner violence, and recent STI symptoms. Participants were asked if they were diagnosed with a STI, would they notify their partner(s)? They were also asked if they thought that their partner(s) would be willing to take medication to treat the STI. Maternal HIV status was determined by rapid HIV antibody testing administered at each visit if the participant was HIV negative at the previous visit.
Women self-collected vulvovaginal swab specimens each study visit, and trained staff tested the swabs for CT, NG, and TV using Xpert® assays (Cepheid, Sunnyvale, CA). Women with a positive STI result were treated in accordance with South Africa National Guidelines. 22 CT infections were treated with 1 g azithromycin orally, NG with an intramuscular injection of 250 mg ceftriaxone plus 1 g azithromycin orally (2 g azithromycin if significant penicillin allergy), and TV with 400 mg metronidazole orally every twelve hours for seven days. In addition to receiving treatment, women were encouraged to notify their partner(s) of their STI diagnosis and given a partner referral letter. The letter included the specific STI(s) that the participant was diagnosed with and treated for along with a recommendation to present to the clinic within seven days for treatment, as well as the work telephone number of the study nurse. Partners had the option of presenting the referral letter to a health care worker at the same clinic, a different clinic, or a private pharmacy to obtain treatment. At the following visit, participants were asked if: (1) they gave their partner(s) the referral letter, (2) their partner(s) went to a clinic or pharmacy for treatment, and (3) their partner(s) took the medication to treat the STI (Figure 1).

Partner notification and treatment for sexually transmitted infections among pregnant women, Cape Town, South Africa, 2017–2019 (n = 242). *Women with an incident STI at the post-partum visit were not included in the partner notification study due to that being the final study visit.
Statistical analysis
This is an analysis from a subset of a cohort study of STI prevalence and incidence in pregnant women. 17 Categorical variables are reported as frequencies and percentages. Continuous variables are presented as medians and interquartile ranges (IQR). We used logistic regression to evaluate factors associated with (1) partner notification and (2) partner notification and treatment. Each adjusted model included continuous maternal age as a potential confounder. Statistical analyses were performed using SAS version 9.4 (SAS Institute, 2013).
Ethics
The Human Research Ethics Committee at the University of Cape Town Faculty of Health Sciences (#454/2017) and the Institutional Review Board at the University of California, Los Angeles (#19–000237) approved the study protocol. Informed consent was obtained from all participants before enrollment.
Results
We enrolled 242 pregnant women at first antenatal visit. The median age of participants was 29 years (IQR = 24–34 years) and median gestational age was 19 weeks (IQR = 13–24 weeks). More than half of participants reported being unmarried, not cohabiting with, or having no relationship with the father of their child (52%, n = 127). Forty-four percent of women were living with HIV at first antenatal visit (n = 107) and 14% reported being in a concordant HIV-infected relationship (n = 34), but 38% did not know their partner’s HIV serostatus (n = 89). The majority of participants reported vaginal sex during pregnancy (93%, n = 225). Three women reported having two or more partners in the past three months (1%). One-third of women suspected their partner of having other sex partners (n = 79). Few women reported intimate partner violence during pregnancy or in the 12 months prior to pregnancy (4%, n = 9). Twenty-one percent reported having ever been diagnosed with a STI prior to the current pregnancy (n = 52). Nearly all women reported that they would notify their partner if they tested positive for a STI in this study (97%, n = 234), and more than three-quarters (78%, n = 189) thought their partner would be willing to take medication to treat the STI (Table 1).
Baseline demographic, clinical and behavioral characteristics of pregnant women attending first antenatal visit, Cape Town, South Africa, 2017–2019 (n = 242).
IQR, interquartile range; STI, sexually transmitted infection; CT, Chlamydia trachomatis; NG, Neisseria gonorrhoeae; TV, Trichomonas vaginalis.
At first antenatal visit, 24% of participants reported recently experiencing STI-related symptoms including abnormal vaginal discharge, increased pain during intercourse, pain during urination, vaginal bleeding, and/or genital sores (n = 59). The most frequently reported symptoms were vaginal discharge (14%, n = 35), followed by pain during urination (6%, n = 15), and pain during intercourse (5%, n = 13) (Supplemental Table 1). Nearly one-third of participants were diagnosed with at least one STI (32%, n = 78). Of those women, 31 (40%) tested positive for CT, two (3%) for NG, 24 (31%) for TV, eight (10%) for CT and NG, nine (12%) for CT and TV, three (4%) for NG and TV, and one (1%) for all three STIs. Among women living with HIV, TV mono-infection (15%, n = 16) was the most common STI, whereas CT mono-infection (15%, n = 20) was more common among HIV-uninfected women (Supplemental Table 2). In addition, two participants were diagnosed with an incident STI at the third trimester visit and three were diagnosed with an incident STI at the post-partum visit. At the third trimester visit, one woman living with HIV tested positive for TV and one HIV-uninfected woman tested positive for CT. At the post-partum visit, three HIV-uninfected women tested positive for CT.
Partner notification and treatment
Overall, 83 women were diagnosed with at least one STI during the study (34%). Partner notification information was not obtainable from the three participants (4%) that tested positive for an incident infection at the post-partum visit due to that being the final study visit. In addition, seven women (8%) were unable to provide answers to the notification-related questions because six did not have a partner during the study period and one was lost to follow-up.
Among the 73 participants who were diagnosed with a STI and had contactable partners, 68 (93%) reported notifying their partner and giving them the referral letter for treatment. Of the five participants (7%) who did not notify their partner, one reported that their partner was out of town, one reported that their partner had relocated to another province, one refused to take the referral letter from the nurse and two reported that their partners refused to take the referral letter, leading to dissolution of the relationship. In addition, three (60%) of these women were living with HIV and two (40%) were HIV-uninfected. Partner notification uptake did not differ by STI diagnosed in the index patient.
Among the 68 women who reported notifying their partner, 47 (69%) reported that their partner sought treatment for the STI and five (7%) did not know if their partner sought treatment. The most common barrier to a partner seeking treatment, as reported by the index patient, was refusal to attend a clinic or pharmacy (44%, n = 7) followed by a conflicting work schedule (31%, n = 5), and a perceived lack of illness due to an asymptomatic infection (6%, n = 1). Three women (19%) did not provide reasons for their partner not seeking treatment.
Of the 47 participants who reported that their partner sought treatment, 46 (98%) reported that their partner took the medication to treat the STI and one (2%) did not know if their partner took the medication. Among the 46 partners treated presumptively for a STI, 12 (26%) partners returned to the same clinic as the index patient (verified by the study nurse). In addition, 12 (26%) women reported that their partner went to a private pharmacy and nine (20%) reported that their partner went to a different clinic. There was no information about where the final 13 (28%) partners were treated. Of the 46 women who reported that their partner was treated, 7 (15%) did not clear their infection and retested positive for at least one of the STIs that they had been diagnosed with at the previous visit.
Factors associated with partner notification
We analyzed results to evaluate factors associated with partner notification following STI diagnosis in pregnant women in antenatal care. Our study was underpowered to detect associations between participant characteristics and partner notification, but nearly all women reporting being married to or cohabiting with the father of their child (96%, n = 27) or recently experiencing STI-related symptoms (96%, n = 25) also reported notifying their partner. All women that reported recently experiencing intimate partner violence (n = 5), being diagnosed with a STI prior to the current pregnancy (n = 11) or being in a concordant HIV-uninfected relationship (n = 23) also reported notifying their partner (Table 2).
Factors associated with partner notification for STIs among pregnant women, Cape Town, South Africa, 2017–2019 (n = 73).
STI, sexually transmitted infection; IQR, interquartile range; OR, odds ratio; CI, confidence interval.
aModels adjusted for maternal age.
Factors associated with partner notification and treatment
Among women with contactable partners (n = 73), 46 (63%) reported both that they notified their partner and their partner took medication to treat the STI, while 27 (37%) either reported that they did not notify their partner or their partner did not take medication to treat the STI. Younger pregnant women aged 18 to 31 years had increased odds of partner notification and treatment (odds ratio [OR] = 3.82; 95% confidence interval [CI] = 1.34–10.90) compared to those aged 32 and older. Seventy-one percent of HIV-uninfected participants notified their partner and reported their partner was treated compared to 56% of women living with HIV, though the HIV status of the participant was not associated with partner notification and treatment (OR = 1.85; 95% CI = 0.70–4.90). Being in a concordant HIV-uninfected relationship was associated with partner notification and treatment in the crude analysis (OR = 2.83; 95% CI = 0.91–8.82) but was attenuated after adjusting for maternal age (aOR = 2.13; 95% CI = 0.61–7.41) (Table 3).
Factors associated with partner notification and treatment for STIs among pregnant women, Cape Town, South Africa, 2017–2019 (n = 73).
STI, sexually transmitted infection; IQR, interquartile range; OR, odds ratio; CI, confidence interval.
aModels adjusted for maternal age.
Discussion
Our study assessed the acceptability of partner notification for STIs among pregnant women in antenatal care at a public clinic in Cape Town, South Africa. We found that almost all women reported that they would notify their partner if they tested positive for a STI, and of those who were diagnosed, the vast majority did report notifying their partner. However, while more than three-quarters of participants thought their partner would be willing to take medication to treat the STI, just under two-thirds (63%) reported that their partner was treated. Our study was underpowered to detect associations between participant characteristics and partner notification, but all women that reported intimate partner violence, being previously diagnosed with a STI or being in a concordant HIV-uninfected relationship reported notifying their partner. Younger women had increased odds of partner notification and treatment. Being in a concordant HIV-uninfected relationship was associated with partner notification and treatment in the crude analysis but was attenuated after adjusting for maternal age.
Similar to a study conducted among pregnant women in Botswana by Offorjebe et al, we found that nearly all participants were willing to notify their partner of a STI diagnosis and just under two-thirds reported that their partner was treated presumptively. 20 However, the proportion of women who reported notifying their partner was higher than that reported in Botswana and also higher than many of the studies identified by a systematic review of partner notification in sub-Saharan Africa. 9 , 20 In addition, the proportion of those reporting partner notification was higher than that previously reported among male and female index patients in Cape Town. 10
In our study, barriers to partner notification included relationship dissolution and partners being out of town, a finding consistent with previous studies conducted in both South Africa and other countries in sub-Saharan Africa. 9 , 12 , 23 Interestingly, while fear of intimate partner violence is a commonly cited barrier, all women that reported recently experiencing intimate partner violence also reported notifying their partner. 11 The most common obstacles to partners being treated were a refusal to visit a clinic or pharmacy and a conflicting work schedule, well-documented barriers to males seeking treatment for both STIs and HIV. 24 , 25 One partner notification strategy that may help address these barriers is expedited partner therapy where the index patient is provided with medication to deliver to their partner, allowing the partner to get treated without visiting a clinic. 13 However, there are potential disadvantages to expedited partner therapy including adverse drug reactions and the inability to screen partners for other STIs and HIV. 26
Prior studies have found that having only one partner, having had a long-term relationship with the partner and considering the partner to be their main partner are predictors of successful partner notification. 27 , 28 Younger pregnant women (ages 18–31) had increased odds of partner notification and treatment compared to older women. However, a study conducted in Louisiana, USA by Kissinger et al. found that women who reported their partner was treated were more likely to be older. 29 Additional studies have found that having only partner, considering that partner to be their main partner and living with the partner are associated with successful partner notification and treatment. 29 , 30
Our study did have some limitations. First, the study was a secondary analysis and was underpowered, which decreased our precision and limited our ability to identify relationships between participant characteristics and partner notification and treatment. Second, our study utilized self-reported responses to collect data on intimate partner violence, sexual behavior, and partner notification and treatment. Thus, intimate partner violence and sexual behavior may be under-reported while partner notification and treatment may be over-reported due to recall bias. Next, the generalizability of our study may be limited because we collected data from one facility. However, we attempted to select a clinic that was representative of others in the region regarding patient socio-demographics (race, gender, and income) and services offered. Finally, this was an analysis of a cohort study that evaluated the prevalence, incidence, and correlates of STIs in pregnant women. As a result, we did not collect data on male partner characteristics, information that would likely be helpful in identifying factors associated with male partner treatment.
Interventions to improve STI partner notification and treatment are urgently required in pregnancy, especially in older women and women living with HIV. In our study, most women reported notifying their partners and over half reported that their partners were treated. In addition, almost all women reported sexual activity during pregnancy. However, we found several women (n = 7, 15%) that were still infected at the subsequent visit, indicating that their partner may not have been treated or treated effectively. Over one-quarter of women reported that their partner was treated in a pharmacy, however, we do not know if they received the appropriate medication to treat the STI, which is especially important for NG infection that is antibiotic resistant and requires an intramuscular injection of ceftriaxone. Future studies are needed on how best to treat male partners of STI-infected women, especially older women and those living with HIV, to ensure that they are treated prior to labor and delivery.
Supplemental Material
sj-pdf-1-std-10.1177_0956462420949789 - Supplemental material for Partner notification and treatment for sexually transmitted infections among pregnant women in Cape Town, South Africa
Supplemental material, sj-pdf-1-std-10.1177_0956462420949789 for Partner notification and treatment for sexually transmitted infections among pregnant women in Cape Town, South Africa by Hunter Green, Sophia Taleghani, Dorothy Nyemba, Landon Myer and Dvora Joseph Davey in International Journal of STD & AIDS
Footnotes
Authors’ contribution
HG and DJD collaborated in the writing of the manuscript. DJD designed and conducted the study and data collection. HG performed the statistical analyses. ST, DN and LM reviewed the manuscript before submission.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: DJD and LM received funding from NIMH (R01MH116771). DJD also received funding from NIH/Fogarty International Center (K01TW011187). Test kits were donated by Cepheid Inc. (Sunnyvale, CA, USA).
Supplemental material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
