Uptake and Persistence of Safer Conception Strategies Among South African Women Planning for Pregnancy

Data for this analysis comes from the ZINK, or Zivikele ngaphambi kokukhulelwa—meaning “Protecting Yourself Before Pregnancy”–study. This single-arm longitudinal safer conception intervention study comprised 330 HIV-uninfected women from KwaZulu-Natal, South Africa planning for pregnancy with a partner with HIV or of unknown serostatus. The primary aim of the ZINK study was to assess oral PrEP uptake and adherence during periconception and pregnancy [20]. Women were eligible to participate if they were 1) aged 18–35 years, 2) were likely to be fertile based on responses to reproductive history assessment, 3) not on a long-acting family planning method, 4) reported personal or partner’s plans for a child within the next 12 months with a 5) partner with HIV or serostatus unknown, 6) fluent in English or isiZulu, and 7) able to provide informed consent [20]. Women were excluded from the study if they were 1) living with HIV, 2) pregnant, 3) hospitalized within 30 days before entry into the study, or receiving treatment for an active illness that could interfere with study participation [20]. Women were enrolled from November 2017 to January 2020 and followed quarterly for a minimum of 12 months. Women who became pregnant during the study were followed through to pregnancy outcomes for a maximum of 24 months. In addition to oral PrEP, women were counseled on safer conception strategies including partner testing and serostatus HIV disclosure, limiting condomless sex to peak fertility, delaying sex without condomless sex to peak fertility, delaying sex without condoms until partner achieves HIV RNA suppression or on antiretroviral therapy (ART) for 6 months for partners with HIV, and waiting to have a child. We also offered couples-based HIV counseling and testing as a safer conception strategy on site, however no woman brought a partner to the site for this service.

Women completed an enrollment questionnaire to assess socio-demographics (including age, education attainment, employment status, marital status, social support [21], and income), reproductive history (including the number of prior pregnancies in a lifetime, any child with the main partner, number of sexual partners, condom use during last sex with the main partner, HIV status of main partner), lifestyle and health-related history (including alcohol consumption, and depression) and other constructs based on a conceptual framework for periconception risk behavior (including HIV knowledge [22], HIV risk perception [23], HIV stigma [24], dyadic trust [25, 26], and sexual relationship power [27]. Safer conception behaviors were assessed at each quarterly visit (every 3 months) through a brief safer conception card which women were asked to complete to indicate their safer conception strategy selections. The safer conception counseling was first offered at the second visit (within 1 month of enrolment), and at each study visit thereafter for those who remained non-pregnant. Women who became pregnant received counseling around HIV risk reduction in the context of pregnancy [20].

Two main outcomes were assessed 1) use of safer conception strategies, which was defined as the proportion of women who self-reported using at least one safer conception strategy during the follow-up period -1, 3, 6, and 9 months, and 2) Self-reported persistent use of safer conception strategy, which was defined as the proportion of women who reported using the same safer conception strategy from 3 months through 9 months follow-up visit.

Women who were not pregnant were asked about the safer conception strategy used at each study visit by research staff. Their strategies were documented on a safer conception card: “Which safer conception strategies is the participant using?” with the following options available: “Wait to have a child” (waiting for a better time to have a child until the risk of HIV acquisition and perinatal transmission are reduced); “Disclosure of HIV status”; “Timed sex without condoms to peak fertility”; “Anti-Retroviral Therapy”; “oral Pre-exposure prophylaxis (oral PrEP)”; “Sperm washing”; “Other”; or “None”. The use of a safer conception strategy was assessed from among the 6 safer conception strategies listed above. The strategies assessed for persistence were those that could be utilized longitudinally including waiting to have a child, timed sex without condoms to peak fertility, or oral PrEP (self-report) from the second visit. Persistence in ART for partners with HIV and sperm washing was not assessed due to the few numbers of people selecting those strategies. Likewise, persistence in HIV disclosure status by partner was not assessed as it is only expected to be a single event. Persistence was assessed among women who indicated they were using any of the three above safer conception strategies at the second visit (within 1 month of enrolment). Women were censored at pregnancy or loss to follow up.

Was assessed by the sexual relationship power scale (SRPS) [27] which includes 23 items that are scored as 2 subscales–relationship (15-item questions) and decision-making dominance (8-item questions). For example, women were asked the following questions about their relationship “If I asked my partner to use a condom, he would get violent”, “Most of the time, we do what my partner wants to do”, “My partner might be having sex with someone else” with 4-choice Likert scale responses. Example questions assessing decision-making dominance included “Who usually has more say about whether you use condoms?” “In general who do you think has more power in your relationship?”, and “Who usually has more say about what types of sexual acts you do?” Response options included “Your partner”, “Both of you equally”, and “You”. A composite score was derived. Possible scores ranged from 1 to 4, with higher scores indicating more power in the relationship in the area of relationship control and decision-making dominance.

Was assessed by the Dyadic Trust Scale [25], consisting of 8-item questions with 5-point Likert scale response options including “My partner is primarily interested in his welfare”, “My partner is perfectly honest and truthful with me”, and “I feel that I can trust my partner completely”, “There are times when my partner cannot be trusted”, and “ I feel that I can trust my partner completely”. Responses to each item were summed to give an overall dyadic trust score. Possible scores ranged from 13 to 40, with higher scores indicating higher levels of dyadic trust.

Was assessed with 24-item questions–15-item questions from the brief HIV questionnaire originally developed by Carey and Schroder [22], and 9-item questions developed by Matthews et al. [15]. For example, women were asked the following questions: “ A person can get HIV by sharing a glass of water with someone who has HIV”, “ A person can get HIV by sitting in a hot tub or swimming pool with a person who has HIV”, “If a woman has HIV then her baby will always be born with HIV”, “If an HIV-positive man has sex without condoms with an HIV-negative woman even once, she will always catch the virus”. Response options included “Yes”, “No”, or “Don’t Know”. An overall HIV score was derived by summing scores of the correct answers. Possible scores ranged from 0 to 24, with higher scores indicating higher knowledge about HIV spread, prevention, and impacts of HIV infection.

Attributed to others was assessed with 12 item questions from the HIV Stigma Scale developed by Visser et al. [24]. For example, women were asked to respond to the following questions with options “I agree” or “I disagree”: “Most people think getting HIV is a punishment for bad behavior”, “Most people think that having HIV is just a matter of bad luck”, “Most people would reject the friendship of someone with HIV”, and “Most people are afraid to be around people with HIV”. Possible scores ranged from 0 to 12, with higher scores indicating a higher level of stigmatizing attitudes attributed to others.

Was assessed with 7 item questions from the Perceived HIV Risk scale originally developed by Napper et al. [23]including “What is your gut feeling about how likely you are to get infected with HIV?”, “I worry about getting infected with HIV”, “I am sure I will not get infected with HIV”, and “I feel I am unlikely to get infected with HIV”. Items were rated on Likert-type scales. Responses to each item were summed, and an overall score was computed. Possible scores ranged from 7 to 34 with higher scores indicating higher perceived HIV risk.

Ethics approval was obtained from the Human Research Ethics Committee at the University of the Witwatersrand (Johannesburg, South Africa), and the Institutional Review Board of Partners Healthcare (Boston, Massachusetts, USA). Women were reimbursed 250 ZAR (~ 20 US$) per study visit [20].

Descriptive statistics were used to summarize the baseline characteristics of study participants. Univariable and multivariable modified Poisson regression models were conducted to assess predictors of persistence on any form of safer conception strategy from 3 months through 9 months. The univariable analysis included the following variables: age (continuous variable), education (Grade 7- Grade 11, Grade 12, and post-grade 12), employment status (employed, not employed), income (< $116, $116-$232), prior pregnancies (0, 1, 2 +), having a child with main partner (yes, no), number of sexual partners past 3 months (0 or 1, 2 +), condom use at last sex with a main partner (yes, no), alcohol consumption (any, never), depression (≥ 1.75, < 1.75), and sexual relationship power, dyadic trust, social support, HIV knowledge, HIV stigma, and perceived HIV risk each treated as a continuous variable. Variables were selected for inclusion in the multivariable analysis based on a p-value of 0.25. The risk ratio with its 95% confidence interval was computed. A p-value of less than 0.05 was considered statistically significant. All analyses were performed using SAS version 9.4.

Table 1 presents the baseline characteristics of eligible study participants. The mean age of women at baseline was 24.6 years. Slightly over a third had completed at least some form of tertiary education–university, college, or vocation (n = 105, 38.6%), while less than a third were employed (n = 71, 26.1%). Nearly half of the women had no pregnancy in their lifetime (n = 124, 45.6%) and twenty-one percent had had a child with their main partner. Most women were in a long-term relationship (for at least 6 months) with their partner (n = 250, 92.3%) and did not know the HIV status of their partner (n = 262, 96.7%).

Figure 1 shows the different safer conception methods and the proportion of women who reported planning to use the methods immediately after the first safer conception counseling (second visit), at 3 months, 6 months, and 9 months. At the second visit (within 1 month of enrolment), the number and proportion e of women who reported selecting the strategies of oral PrEP, timed sex without a condom to peak fertility, waiting to have a child, disclosure of HIV status, ART for an infected partner, or sperm washing were 129 (50%), 86 (33%), 27 (10%), 20 (8%), 1 (0.4%), and 0 (0%), respectively. At 3, 6, and 9 months, the number and proportion of women who reported using oral PrEP was: 130 (55%), 116 (54%), and 103 (52%); timed sex without condom to peak fertility – 80 (34%), 90 (42%), and 83 (42%); waiting to have a child – 16 (7%), 10 (4%), and 11 (6%); disclosure of HIV status – 8 (3%), 4 (2%), and 5 (3%) respectively. No woman reported ART for a partner with HIV as a safer conception strategy at the 3-, 6-, or 9-month follow-up visit, while only one woman reported sperm washing as a safer conception strategy utilized at the 9-month follow-up visit.

Table 2 shows the proportion of women who persisted in the use of selected safer conception strategies from initiation at the second visit through 3 months, 6 months, and 9 months of follow-up: oral PrEP, timed sex without a condom to peak fertility, and waiting to have a child. The proportion of women who reported persisting with oral PrEP was 97% (109 of 112 women), 92% (85 of 92 women), and 86% (70 of 81 women) through 3,6, and 9 months, respectively. The proportion of women who reported persisting with timed sex without a condom to peak fertility was 84% (63 of 75 women), 76% (54 of 71 women), and 76% (48 of 63 women) through 3, 6, and 9 months, respectively. The proportion of women who reported persisting with waiting to have a child was 30%, 12%, and 12% through 3, 6, and 9 months respectively.

Among women who did not continue a selected strategy, we describe what methods, if any, women reported using subsequent to discontinuation of the strategy at months 3,6 and 9 of follow-up (Table 3). The proportion of women who reported switching from oral PrEP to another safer conception strategy or using another strategy was 0% at 3 months, 7% at 6 months, and 12% at 9 months. The proportion of women who reported switching from timed sex without a condom to peak fertility to another safer conception strategy or using another strategy was 16% at 3 months, 20% at 6 months and 21% at 9 months. The proportion of women who reported switching from waiting to have a child to another safer conception strategy or using another strategy was 66% at 3 months, 80% at 6 months, and 71% at 9 months.

Table 3 depicts Poisson regression models predicting persistence in one safer conception strategy from three months through 9 months. The percentage of women who reported persisting in any form of safer conception strategy through 9 months was 74.2%. Women who were currently employed were significantly more likely to report persisting in any form of safer conception method (RR = 1.30, 95% CI: 1.07–1.59, p = 0.01). Also, HIV knowledge was positively associated with persisting in any form of safer conception method (RR = 1.04, 95% CI: 1.01–1.08, p = 0.02). Persistence in the use of any form of safer conception method did not differ by other factors including age, income, alcohol consumption, number of sexual partners, sexual relationship power, HIV stigma, or perceived HIV risk.