http://orcid.org/0000-0002-6280-2290
Figures
Abstract
Background
Pregnancy incidence rates among women living with HIV (WLWH) have increased over time due to longer life expectancy, improved health status, and improved access to and HIV prevention benefits of combination antiretroviral therapy (cART). However, it is unclear whether intended or unintended pregnancies are contributing to observed increases.
Methods
We analyzed retrospective data from the Canadian HIV Women’s Sexual and Reproductive Health Cohort Study (CHIWOS). Kaplan-Meier methods and GEE Poisson models were used to measure cumulative incidence and incidence rate of pregnancy after HIV diagnosis overall, and by pregnancy intention. We used multivariable logistic regression models to examine independent correlates of unintended pregnancy among the most recent/current pregnancy.
Results
Of 1,165 WLWH included in this analysis, 278 (23.9%) women reported 492 pregnancies after HIV diagnosis, 60.8% of which were unintended. Unintended pregnancy incidence (24.6 per 1,000 Women-Years (WYs); 95% CI: 21.0, 28.7) was higher than intended pregnancy incidence (16.6 per 1,000 WYs; 95% CI: 13.8, 20.1) (Rate Ratio: 1.5, 95% CI: 1.2–1.8). Pregnancy incidence among WLWH who initiated cART before or during pregnancy (29.1 per 1000 WYs with 95% CI: 25.1, 33.8) was higher than among WLWH not on cART during pregnancy (11.9 per 1000 WYs; 95% CI: 9.5, 14.9) (Rate Ratio: 2.4, 95% CI: 2.0–3.0). Women with current or recent unintended pregnancy (vs. intended pregnancy) had higher adjusted odds of being single (AOR: 1.94; 95% CI: 1.10, 3.42), younger at time of conception (AOR: 0.95 per year increase, 95% CI: 0.90, 0.99), and being born in Canada (AOR: 2.76, 95% CI: 1.55, 4.92).
Conclusion
Nearly one-quarter of women reported pregnancy after HIV diagnosis, with 61% of all pregnancies reported as unintended. Integrated HIV and reproductive health care programming is required to better support WLWH to optimize pregnancy planning and outcomes and to prevent unintended pregnancy.
Citation: Salters K, Loutfy M, de Pokomandy A, Money D, Pick N, Wang L, et al. (2017) Pregnancy incidence and intention after HIV diagnosis among women living with HIV in Canada. PLoS ONE 12(7): e0180524. https://doi.org/10.1371/journal.pone.0180524
Editor: Ruanne V. Barnabas, University of Washington Department of Global Health, UNITED STATES
Received: January 5, 2017; Accepted: June 17, 2017; Published: July 20, 2017
Copyright: © 2017 Salters et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript.
Funding: CHIWOS is funded by the Canadian Institutes of Health Research (CIHR, MOP111041); the CIHR Canadian HIV Trials Network (CTN 262); the Ontario HIV Treatment Network (OHTN); and the Academic Health Science Centres (AHSC) Alternative Funding Plans (AFP) Innovation Fund. KS and AC received doctoral scholarship support from the Canadian Institutes of Health Research (CIHR). AdP received support from Fonds de Recherche du Quebéc – Santé (FRQS) (Chercheur-boursier clinicien – Junior. NO received support from FRQS. AK received salary support through a Tier 2 Canada Research Chair in Global Perspectives on HIV and Sexual and Reproductive Health. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Health and survival outcomes for people living with HIV have improved dramatically since the advent of combination antiretroviral therapy (cART) [1–3]. HIV treatment with sustained viral suppression is also enabling safer reproductive options for women living with HIV (WLWH), including better maternal health, improved fertility [4], negligible risk of HIV transmission to partners during condomless sex [5–7], and dramatic reductions in perinatal HIV transmission risk [8–10]. These improvements have transformed the reproductive health landscape for people living with or affected by HIV [11–17], with studies demonstrating that WLWH are more likely to become pregnant and have children in the modern cART era than in earlier years of the HIV epidemic [18, 19].
Among WLWH in Canada, cART use is associated with increased fertility intention and an increase in the proportion of WLWH becoming pregnant over time [12]; a finding that has been corroborated in other settings [15, 18–23]. However, it is unknown whether such increases in pregnancy incidence are driven by intended or unintended pregnancies [13, 15, 24–26], raising questions about how women’s reproductive health and agency are prioritized and supported after an HIV diagnosis.
Supporting WLWH to plan for desired pregnancy and prevent unintended pregnancy is a critical component of comprehensive women-centred HIV care [27] and central to women’s sexual and reproductive health and rights and HIV prevention efforts [8, 28–32]. Unintended pregnancies have numerous potential health implications including, higher risks of maternal and infant morbidity and mortality, delayed engagement in antenatal care, poorer birth outcomes, and poorer retention in postpartum follow-up [33]. Such risks are exacerbated for WLWH, for whom early engagement in pre-conception, antenatal, and postpartum care is key to maximizing maternal health and minimizing risks of perinatal transmission [34]. A pregnancy after an HIV diagnosis can be a stressful time, with women expressing significant worry and fear about HIV-related concerns (i.e., their own health and the health of their infant) as well as non HIV-related concerns (i.e., financial and relationship stress), coupled with experiences of internalized and experienced HIV-related stigma around becoming pregnant [35, 36]. Determining the contribution of unintended pregnancy to the overall incidence of pregnancy among WLWH is a key step towards the development of comprehensive reproductive rights- and evidence-based programming across the continuum of reproductive care including pre-conception care when pregnancy is intended, contraceptive access when it’s not, and the availability of safe termination services.
The primary objective of this study was to measure cumulative incidence and incidence rate of pregnancy after HIV diagnosis among a cohort of WLWH in Canada overall and by pregnancy intention, cART use status at time of pregnancy, and cART era (before or after the year 2000). Our secondary objective was to determine independent socio-demographic and clinical correlates of unintended pregnancy among the most recent or current pregnancy. This information is key to the provision of targeted and tailored reproductive health services for WLWH, to mitigate health risks, support reproductive rights, and focus prevention efforts more effectively.
Methods
Study setting
As of 2011, there were approximately 16,600 WLWH in Canada, an increase of nearly 13% from three years prior [37]. The majority of WLWH reside in the three Canadian provinces of British Columbia, Ontario and Quebec. HIV prevalence and incidence are inequitably distributed among women in Canada along several social axes, including poverty, Indigenous ancestry, injection drug use and/or sex work histories, refugee and newcomer status, African, Caribbean, or Black Canadian (ACB) ethnicity, and sexual and gender identities, with several points of intersection between and within these groups [37].
Study design and guiding frameworks
This analysis used baseline questionnaire data from the Canadian HIV Women’s Sexual and Reproductive Health Cohort Study (CHIWOS), a multi-site, community-based research study conducted by, with, and for WLWH in collaboration with researchers, healthcare providers, policy-makers, and other stakeholders. As described in detail elsewhere [27, 38], the primary objectives of CHIWOS are to assess the prevalence, barriers, and facilitators to use of women-centred HIV care, and the impact of such patterns of use on health outcomes. CHIWOS is grounded in community-based research principles [39], and guided by Social Determinants of Health [40, 41] and Critical Feminism [42] frameworks, with the aim of producing community-driven research and action on health priorities for WLWH.
Study population, recruitment, and procedures
Eligible participants included WLWH (self-identified) aged ≥16 years, including trans and gender diverse women, residing in British Columbia, Ontario, and Quebec. Between August 27, 2013 and May 1, 2015, women were recruited to participate in CHIWOS through peer word-of-mouth, HIV clinics, AIDS Service Organizations, non-HIV community-based organizations, the networks of our national Steering Committee and three provincial Community Advisory Boards, and online methods such as listservs for WLHIV and our study’s website (www.chiwos.ca), Facebook (www.facebook.com/CHIWOS), and Twitter (www.twitter.com/CHIWOSresearch) pages [43]. At enrolment, participants completed structured, comprehensive questionnaires in English or French using online FluidSurveys™ software [44], administered by Peer Research Associates, who are WLWH who completed training in community-based research methods [45]. Study visits took 1.5 to 2.5 hours and were completed either in-person at collaborating HIV clinics, AIDS Service Organizations, or women’s homes, or via phone/Skype for those living in rural and remote areas. Follow-up interviews at 18-month intervals are in-progress.
Participants provided voluntary, informed consent either by signing a consent form or, if questionnaires were completed remotely, via verbal consent with a witness. Ethical approval for this study was obtained from the Research Ethics Boards of Simon Fraser University, University of British Columbia/Providence Health, Women’s College Hospital, and McGill University Health Centre.
Analytic sample: Inclusion and exclusion criteria
A total of 1,424 WLHIV enrolled in the CHIWOS study and completed the baseline visit. Baseline questionnaire data were used for this analysis including reports of lifetime history of pregnancy (up to eight pregnancy events in a lifetime). For this analysis, we excluded participants who did not identify their biological sex as female (n = 57), women aged 45 or older at HIV diagnosis (n = 139), women who reported completing menopause before HIV diagnosis (n = 16), and women with unknown HIV diagnosis year (n = 47), yielding an analytic sample of 1,165 women. We further restricted pregnancy incidence analyses to women who were diagnosed with HIV prior to date of pregnancy (estimated at conception). For pregnancy incidence analyses and time-to-event calculations, we also excluded 19 participants missing pregnancy dates and 10 who reported a pregnancy end date that was the same as the HIV diagnosis date. These exclusions yielded an analytic sample of 1,136 women. Finally, for the analysis of correlates of unintended pregnancy, we restricted the analysis to the most recent or current pregnancy reported after HIV diagnosis.
Measures
Primary outcome: Pregnancy incidence after HIV diagnosis.
We assessed total number of lifetime pregnancies and detailed retrospective event-level data for all reported pregnancies, up to a maximum of eight lifetime pregnancies. The primary outcome was all self-reported pregnancy events occurring after HIV diagnosis until censored (CHIWOS interview date), including current pregnancies. For each pregnancy event reported after HIV diagnosis, data were captured on pregnancy outcome (single live birth, multiple live births, miscarriage, still birth, pregnancy termination, or ectopic pregnancy), pregnancy intention (unintended vs. intended vs. don’t know), HIV status of the other biological parent (HIV positive vs. HIV negative vs. unknown), on cART before pregnancy occurred (yes vs. no), date of pregnancy outcome (month and year), duration of pregnancy, and, if applicable, infant HIV status (HIV-positive, HIV-negative, testing underway, or unknown). We estimated date of conception by subtracting the duration of pregnancy from the date of pregnancy outcome.
To assess the pregnancy incidence rate, we applied the following rules to calculate woman-years (WY) ‘at risk’ for pregnancy: (1) Time of follow-up began at reported HIV diagnosis date or at 16 years of age, whichever date was later to avoid biasing time at risk among perinatally infected women and (2) Women were censored at age 45, date of surgically induced menopause (hysterectomy), or date of CHIWOS interview, whichever occurred first.
For analyses focused on the most recent or current pregnancy, we assessed additional information regarding the pregnancy experience, including ‘feeling when found out about the current or most recent pregnancy’ (very happy/happy vs. not sure/unhappy/very unhappy vs. unknown), ‘happiness during current or most recent pregnancy’ (happiest time/happy time vs. moderately hard/hard/worst times vs. unknown) [46]. Considering these additional data, we only captured most recent or current pregnancy in the bivariate and multivariable analyses.
Explanatory variables.
We examined associations of pregnancy incidence after HIV diagnosis and unintended pregnancy with baseline variables, identified a priori from the established literature [11, 12, 20, 23, 25, 26]. Socio-demographic variables included: age at interview [median and interquartile range (IQR)], province of interview (British Columbia vs. Ontario vs. Quebec), ethnicity (Indigenous vs. African/Caribbean/Black vs. White vs. other ethnicities), whether the participant was born in Canada (yes vs. no), legal relationship status at interview (married/common-law vs. single), sexual orientation [heterosexual vs. lesbian, gay, bisexual, two-spirit or queer (LGBTQ)], education (<high school vs. ≥high school), annual personal income at interview (<$20,000 vs. ≥$20,000 per year), history of injection drug use (yes vs. no) and history of incarceration (yes vs. no).
Health-related variables included Hepatitis C (HCV) co-infection (yes vs. no) and lifetime diagnosis of any mental health condition (yes vs. no). HIV clinical variables included current cART use (current use vs. previous use vs. never) and years living with HIV at interview. We defined cART initiation era as ‘before the year 2000’ or ‘2000 or later’ as this is a general mark of when therapy guidelines shifted to include evidence on the benefits of cART [47, 48].
Statistical analysis
For the primary objective, we assessed baseline characteristics of the sample of 1,165 WLWH including pregnancy event information, demographics, and HIV diagnosis information. We used Kaplan-Meier (K-M) methods to estimate cumulative incidence of time to first pregnancy after HIV diagnosis or age 16 (whichever came later) to CHIWOS interview date. Overall cumulative incidence of first pregnancy after HIV diagnosis was calculated as well as by reported pregnancy intention (intended vs. unintended), by cART status (on cART vs. not on cART prior to the start of pregnancy), and by cART era (prior to the year 2000 vs. the year 2000 until 2015). Log-rank test assessed statistical differences in the K-M curves.
Generalized estimating equation (GEE) Poisson models were used to calculate pregnancy incidence rates expressed as incidence rate per 1,000 women-years (WY) with 95% confidence intervals overall, by pregnancy intention, and cART status.
For the secondary objective, baseline characteristics of study participants were compared for women with reported recent or current pregnancy by pregnancy intention, using Pearson χ2 test or Fisher’s exact test for categorical variables and Wilcoxon’s Rank Sum test for continuous variables. Multivariable logistic regression was used to identify independent correlates of unintended pregnancy of the most recent or current pregnancy. Model selections were conducted using a backward stepwise elimination technique based on two criteria (Akaike Information Criterion (AIC) and Type III p-values).
Results
Baseline characteristics of study population
Among 1,165 WLWH included in the analysis, 30.1% identified as African/Caribbean/Black, 22.2% identified as Indigenous, 40.9% identified White and 6.8% as other ethnicities; 24.4% were from British Columbia, 51.5% from Ontario, and 24.1% from Quebec. Our sample included 12.2% LGBTQ women, 30.7% with a history of injection drug use, and 35.7% born outside of Canada. Median age at interview was 41 (IQR: 35–48) and median years since HIV diagnosis was 12 (IQR: 6–17).
Pregnancy incidence and outcomes
Among 1,165 WLWH, 278 (23.9%) women reported a total of 492 pregnancies after being diagnosed with HIV, with 299 (60.8%) of those pregnancies reported as unintended. Of 492 pregnancies, 92 (18.7%) resulted in miscarriage or stillbirth, 101 (20.5%) were terminated, and 282 (57.3%) resulted in a single or multiple live birth. Seventeen women (3.5%) were pregnant at time of interview. Among the 282 live births, 94.7% of infants tested HIV-negative (Table 1). Among the sample of 1,165 WLWH, the median number of pregnancies from time of HIV diagnosis until interview, including miscarriage, terminations and current pregnancies, was 2 (IQR: 1–4) and the median number of children was 1 (IQR: 0–3). Of the pregnancies occurring after HIV diagnosis, the median age at conception was 30 years [IQR: 25–34] and in 328 (66.7%) of pregnancies, women had initiated cART prior to the start of pregnancy.
Using a further restricted sample of 1,136 with HIV diagnosis and pregnancy dates provided, we identified that by 1, 5, 10, and 20 years post HIV diagnosis, the cumulative probability of first pregnancy after HIV diagnosis was 3%, 17%, 27% and 32%, respectively (Fig 1). Unintended pregnancy (vs. intended) was cumulatively higher over time; with K-M curves diverging at 1-year post HIV diagnosis (2.4% vs. 0.9%, p = 0.007), at 5 years post HIV diagnosis (11.4% vs. 6.5%, p<0.001), and at 10 years post HIV diagnosis (16.8% vs 11.7%, p = 0.001) (Fig 2). At the 20-year mark after HIV diagnosis, the cumulative incidence of first pregnancy for those with unintended pregnancy was 19.2% (95% CI: 16.2%-22.6%) compared to pregnancy incidence among those with intended pregnancy of 14.6% (95% CI: 11.8%-18.0%). Women on cART prior to pregnancy had higher probability of pregnancy over time; with K-M curves diverging at 5 years post HIV diagnosis (10.9% vs. 6.9%, p = 0.004) and at 10 years post HIV diagnosis (19.3% vs. 8.9%, p<0.001) (Fig 3). At the 20-year mark post HIV diagnosis, the cumulative incidence of first pregnancy for those on cART was 23.8% (95% CI: 20.4%-27.8%) compared to pregnancy incidence among those not on cART at time of pregnancy with prevalence of 9.2% (95% CI: 7.4%-11.6%). Lastly, while there was no difference at the 1- or 5-year mark, women who initiated cART before the year 2000 had a significantly lower probability of pregnancy at the 10-year and 20-year mark. At the 20-year mark post HIV diagnosis, the cumulative incidence of first pregnancy for those who initiated cART before the year 2000 was 26.7% (95% CI: 21.1%-33.5%) compared to those initiated therapy in the year 2000 or later (39.2% [95%CI: 33.8%-45.1%]) (Fig 4).
This figure demonstrates the estimated time to first pregnancy event (inclusive of live births, terminations and miscarriages/stillbirths) for all women living with HIV in our sample that provided HIV diagnosis date and estimated date of pregnancy (n = 1,136). This figure demonstrates the cumulative incidence estimates and 95% confidence intervals at 1-, 5-, 10-, and 20-years for any pregnancy event.
This figure demonstrates the estimated time to first pregnancy event (inclusive of live births, terminations and miscarriages/stillbirths) for all women living with HIV in our sample that provided HIV diagnosis date and estimated date of pregnancy (n = 1,136). These pregnancies are stratified by self-reported intention as either intended or unintended pregnancy. This figure demonstrates the cumulative incidence estimates and 95% confidence intervals at 1-, 5-, 10-, and 20-years for unintended and intended pregnancies.
This figure demonstrates the estimated time to first pregnancy event (inclusive of live births, terminations and miscarriages/stillbirths) for all women living with HIV in our sample that provided HIV diagnosis date and estimated date of pregnancy (n = 1,136). These pregnancies are stratified by self-reported uptake of cART prior to the start of pregnancy. This figure demonstrates the cumulative incidence estimates and 95% confidence intervals at 1-, 5-, 10-, and 20-years for pregnancies cART exposed and those pregnancies not exposed to cART.
This figure demonstrates the estimated time to first pregnancy event (inclusive of live births, terminations and miscarriages/stillbirths) for all women living with HIV in our sample that provided HIV diagnosis date and estimated date of pregnancy (n = 1,136). These pregnancies are stratified by self-reported date of cART initiation (prior to the year 2000 vs the year 2000 or later). This figure demonstrates the cumulative incidence estimates and 95% confidence intervals at 1-, 5-, 10-, and 20-years for all pregnancies by cART era.
After excluding pregnancies with unknown delivery/termination date, 256/1,146 women reported 423 pregnancies after HIV diagnosis (incidence = 41.6 per 1,000 WYs; 95% CI: 36.6–47.2) over 10,172.5 WYs of follow-up (Table 2). Incidence of unintended pregnancy (24.6 per 1,000 WYs; 95% CI: 21.0–28.7) was significantly higher than the incidence of intended pregnancy (16.6 per 1,000 WYs; 95% CI: 13.8–20.1) after HIV diagnosis. By cART use, pregnancy incidence among WLWH on cART before or during pregnancy (29.1 per 1000 WYs; 95% CI: 25.1–33.8) was significantly higher than the pregnancy incidence among WLWH not on cART during pregnancy (11.9 per 1000 WYs; 95% CI: 9.5–14.9).
Correlates of most recent/current pregnancy after HIV diagnosis
Of the 265 most recent or current pregnancies, 110 (41.5%) were intended and 155 (58.5%) were unintended (Table 3), with only 62 (23.4%) pregnancies occurring prior to the year 2000 (pre-cART era). Women who reported their recent/current pregnancy as unintended compared to intended, were significantly (p<0.05) more likely to identify as Indigenous (vs White) (20.0% vs. 11.8%), born in Canada (63.9% vs. 39.1%), be currently single (60.0% vs. 45.5%), co-infected with HCV (38.7% vs. 21.8%), have a history of incarceration (45.8% vs. 26.4%), have a history of injection drug use (38.7% vs. 23.6%), initiated ART before the year 2000 (29.0% vs. 15.5%), and to report having been previously diagnosed with a mental health disorder (inclusive of depression, anxiety, post-traumatic stress disorder and addiction) (49.0% vs. 32.7%).
With respect to pregnancy experiences, women with an unintended vs. intended recent/current pregnancy were more likely to report feeling unhappy about the referent pregnancy (56.1% vs. 7.3%) and identifying the pregnancy experience as ‘one of the worst times of [her] life’ (60.0% vs. 16.4%). Women who reported a recent/current pregnancy as unintended vs. intended were also more likely to report pregnancy termination (28.4% vs. 1.8%).
In multivariable analyses (Table 4), factors independently associated with a most recent/current unintended pregnancy included being currently single (AOR: 1.94, 95% CI: 1.10, 3.42), being younger at time of pregnancy (AOR: 0.95 per year increase, 95% CI: 0.90, 0.99), and being born in Canada (vs outside of Canada) (AOR: 2.76, 95% CI: 1.55, 4.92).
Discussion
Among women living with HIV in Canada, we found that approximately one in four women reported pregnancy after HIV diagnosis and that pregnancy was more frequent in the modern cART era and with uptake of cART. Pregnancy incidence rates observed in this study are lower than rates observed in other studies of WLWH in other global settings [11, 15, 49, 50], however is consistent with a lower pregnancy rate in the general Canadian population (54.6 per 1000 women) [51].
Approximately 60% of pregnancies after HIV diagnosis among women in this study were reported as unintended and the prevalence varied by key socio-demographic characteristics. The high observed prevalence of unintended pregnancy is consistent with estimates from a cohort of WLWH in Ontario, where 57% of women reported that their most recent pregnancy was unplanned [12] and comparable to the high rates of unintended pregnancies observed among WLWH in other high-resource settings [52, 53]. Research conducted in Sub-Saharan Africa, including Botswana and South Africa demonstrate that 37% to 69% of pregnancies among WLWH are unintended [20, 54–56]. In the United States, approximately 69% of a total of 620 pregnancy events among a sample of WLWH from 26 different regions were determined to be unplanned [53].
This study is the first of its kind to look at the impact of cART uptake and pregnancy rates among WLWH in Canada. After decades of declining numbers of women becoming pregnant after their HIV diagnosis [29, 36], we observed that WLWH on cART are more likely to become pregnant compared to those not on cART prior to pregnancy, with incidence rates nearly three times of that observed among WLWH not on cART. It must be noted that the majority of pregnancy events in our sample (86.5%) occurred after the year 2000, and these WLWH were more likely to become pregnant, and sooner after HIV diagnosis, compared to women in the earlier treatment era. It is important to note that while a previous Canadian study found no association between use of cART and fertility intentions, our study shows an association between cART use and pregnancy events, further highlighting the importance of addressing unintended pregnancy rates for all WLWH [17].
Previous studies have shown increases in pregnancy incidence after cART initiation [15, 20, 23, 57]. Observed increases in pregnancies for WLWH on cART are likely due to a combination of behavioural and biological factors including, increased life expectancy, increase in sexual activity, and improved health and fertility status [31, 58, 59]. While some research has stated that uptake of cART may improve engagement in care and thus increase uptake of contraceptives [60], our data suggests that a gap in family planning specific to WLWH exists given that 60% of pregnancies are considered unintended. Our data indicate that pregnancy is more common with cART uptake and current trends suggest future increases, although the role of contraceptive uptake will need to be more thoroughly investigated. While we did not capture lifetime contraceptive use patterns, data on contraceptive use at baseline among sexually active women enrolled in CHIWOS reveal that 27.4% had not used an effective contraceptive method in the previous six months [61]. Among those women who did report using contraception, nearly half relied on the male condom [61]. This emerging work, as well as our finding around the high prevalence of unintended pregnancy, underscores the need for improved integration of family planning care as part of comprehensive HIV care for women [11, 62, 63].
Our study suggests that more than half of pregnancies among WLWH are unintended similar to rates observed in other studies of WLWH, which range from 51–90% [32, 34]. Such rates are significantly higher than recent estimates among Canadian women of reproductive age in the general population, among whom an estimated 27% of pregnancies are unintended [64]. Pregnancy termination occurred in one in five pregnancies among WLWH in our sample. Consistent with earlier findings, women in our sample were significantly more likely to terminate an unintended pregnancies compared to a reported intended pregnancy [20, 31, 65].
There are several limitations to this study. Our analyses rely on data collected via questionnaire suggesting recall bias is likely present. The retrospective data collection on lifetime history of pregnancy did not allow for an assessment of important behaviours prior to or during pregnancy including contraceptive use at the time of each reported pregnancy, relationship status, smoking, substance use and other socio-behavioural factors. Moreover, with a sensitive outcome of interest being collected retrospectively, there may be some misclassification bias due to shifting perceptions of pregnancy intention in relation to the subsequent outcome of that pregnancy. However, research has suggested that retrospectively collecting data on pregnancy intention likely does not lead to misleading estimates of the prevalence of unintended pregnancies [66], however this has not been validated in a context with WLWH where there may still be significant stigma attached to becoming pregnant or intending to become pregnant. In an attempt to mitigate the influence of bias, questionnaires were administered by trained peer research associates, who were also living with HIV, and were able to guide participants through their pregnancy history. The authors believe this may better support accurate data collection (relative to self-administered or non-peer administered surveys). As well, participants were aware that they were being interviewed by women who shared a lived experience of living with HIV, which likely reduced social desirability bias.
Our rates of pregnancy may be underestimating fertility rates of WLWH as our cohort is older and may not represent all women who are currently living with HIV of reproductive age. Moreover, post-menopausal women who were under 45 years of age were not appropriately censored in this analysis since we had no date of spontaneous menopause As such, our at-risk time may be inflated and our estimates of pregnancy incidence may be lower than actual rates. Although we excluded 127 women who were 45 years old or older at the time of HIV diagnosis, censoring women at 45 was deliberately done to most accurately capture those likely to become pregnant. We found only fewer than 5 women in our sample over the age of 45 who had pregnancy events after HIV diagnosis. In addition, we did not have detailed pregnancy data after eight lifetime pregnancy events and 28 (2.4%) of WLWH in our sample reported more than eight pregnancies. Despite these limitations, our study provides the most comprehensive picture of pregnancy incidence among WLWH in Canada with over 1400 participants and on-going data collection that providing longitudinal data on sexual and reproductive health outcomes among WLWH.
Given known risks of unintended pregnancy to maternal, partner, and perinatal health, our findings underscore the need for comprehensive women-centered sexual and reproductive health programming for WLWH, including provision of effective contraceptive options and pregnancy planning support, and access to safe termination services. In the modern cART era, WLWH must have access to comprehensive reproductive health care and support that extends beyond concerns of sexual and perinatal HIV transmission. These changing reproductive health patterns, including more WLWH becoming pregnant and a high prevalence of unintended pregnancy, have significant implications for women’s overall well-being. Promotion of reproductive health care, supporting women’s reproductive autonomy, decisions and desires, and enhancing access to effective contraceptive options alongside safe, non-coercive pregnancy termination services should be available in conjunction with evolving HIV care and improved therapeutic options for WLWH.
Acknowledgments
The Canadian HIV Women’s Sexual and Reproductive Health Cohort Study (CHIWOS) Research Team would like to especially thank all of the women living with HIV who participate in this research. We also thank the entire national team of Co-Investigators, Collaborators, and Peer Research Associates. We would like to acknowledge the national Steering Committee, the three provincial Community Advisory Boards, the national CHIWOS Aboriginal and African, Caribbean and Black Advisory Boards, and our partnering organizations for supporting the study, especially those who provide interview space and support to our Peer Research Associates. We dedicate this paper to the memory of Marisol Desbiens, a champion for the sexual and reproductive rights of women living with HIV.
Author Contributions
- Conceptualization: KS ML ADP KW TC DD AK DM NP.
- Data curation: KS AC LW SJ AK.
- Formal analysis: KS AC LW SJ AK.
- Funding acquisition: ML, ADP, AK.
- Investigation: ML ADP AC KW TC DD NO KPB AK.
- Methodology: KA ML ADP LW AK.
- Project administration: KA ML ADP DM NP AC NO KPB AK.
- Resources: AK ML ADP.
- Supervision: ML ADP DM NP AK.
- Visualization: KS LW AK.
- Writing – original draft: KS AK.
- Writing – review & editing: KS ML ADP DM NP LW SJ AC KW TC DD NO KPB AK.
References
- 1. Hogg RS, Yip B, Kully C, Craib KJ, O'shaughnessy MV, Schechter MT, Montaner JS. Improved survival among HIV-infected patients after initiation of triple-drug antiretroviral regimens. Canadian Medical Association Journal. 1999 Mar 9;160(5):659–65. pmid:10102000
- 2. Hogg RS, Yip B, Chan KJ, Wood E, Craib KJ, O'shaughnessy MV, Montaner JS. Rates of disease progression by baseline CD4 cell count and viral load after initiating triple-drug therapy. Jama. 2001 Nov 28;286(20):2568–77. pmid:11722271
- 3. Patterson S, Cescon A, Samji H, Chan K, Zhang W, Raboud J, et al. Life expectancy of HIV-positive individuals on combination antiretroviral therapy in Canada. BMC infectious diseases. 2015 Jul 17;15(1):274.
- 4. Haddad LB, Wall KM, Mehta CC, Golub ET, Rahangdale L, Kempf MC, et al. Trends of and factors associated with live-birth and abortion rates among HIV-positive and HIV-negative women. American journal of obstetrics and gynecology. 2017 Jan 31;216(1):71–e1. pmid:27640942
- 5. Cohen MS, Chen YQ, McCauley M, Gamble T, Hosseinipour MC, Kumarasamy N, et al. Prevention of HIV-1 infection with early antiretroviral therapy. New England journal of medicine. 2011 Aug 11;365(6):493–505. pmid:21767103
- 6. Rodger AJ, Cambiano V, Bruun T, Vernazza P, Collins S, Van Lunzen J, et al. Sexual activity without condoms and risk of HIV transmission in serodifferent couples when the HIV-positive partner is using suppressive antiretroviral therapy. Jama. 2016 Jul 12;316(2):171–81. pmid:27404185
- 7.
Rodger A, Bruun T, Cambiano V, Vernazza P, Strada V, Van Lunzen J. 153LB: HIV transmission risk through condomless sex if HIV+ partner on suppressive ART: PARTNER Study. In21st Conference on Retroviruses and Oppotunistic Infections 2014 Mar (pp. 3–6).
- 8. Cooper ER, Charurat M, Mofenson L, Hanson IC, Pitt J, Diaz C, et al. Combination antiretroviral strategies for the treatment of pregnant HIV-1-infected women and prevention of perinatal HIV-1 transmission. JAIDS-HAGERSTOWN MD-. 2002 Apr 15;29(5):484–94.
- 9. Darak S, Darak T, Kulkarni S, Kulkarni V, Parchure R, Hutter I, Janssen F. Effect of highly active antiretroviral treatment (HAART) during pregnancy on pregnancy outcomes: experiences from a PMTCT program in western India. AIDS patient care and STDs. 2013 Mar 1;27(3):163–70. pmid:23477457
- 10. Okafor II, Ugwu EO, Obi SN, Odugu BU. Virtual Elimination of Mother-to-Child Transmission of Human Immunodeficiency Virus in Mothers on Highly Active Antiretroviral Therapy in Enugu, South-Eastern Nigeria. Annals of medical and health sciences research. 2014;4(4):615–8. pmid:25221715
- 11. Kaida A, Matthews LT, Kanters S, Kabakyenga J, Muzoora C, Mocello AR, et al. Incidence and predictors of pregnancy among a cohort of HIV-positive women initiating antiretroviral therapy in Mbarara, Uganda. PloS one. 2013 May 21;8(5):e63411. pmid:23704906
- 12. Loutfy MR, Hart TA, Mohammed SS, Su D, Ralph ED, Walmsley SL, et al. Fertility desires and intentions of HIV-positive women of reproductive age in Ontario, Canada: a cross-sectional study. PloS one. 2009 Dec 7;4(12):e7925. pmid:19997556
- 13. Maier M, Andia I, Emenyonu N, Guzman D, Kaida A, Pepper L, et al. Antiretroviral therapy is associated with increased fertility desire, but not pregnancy or live birth, among HIV+ women in an early HIV treatment program in rural Uganda. AIDS and Behavior. 2009 Jun 1;13:28–37. pmid:18389364
- 14. Mmbaga EJ, Leyna GH, Ezekiel MJ, Kakoko DC. Fertility desire and intention of people living with HIV/AIDS in Tanzania: a call for restructuring care and treatment services. BMC public health. 2013 Jan 30;13(1):86.
- 15. Myer L, Carter RJ, Katyal M, Toro P, El-Sadr WM, Abrams EJ. Impact of antiretroviral therapy on incidence of pregnancy among HIV-infected women in Sub-Saharan Africa: a cohort study. PLoS medicine. 2010 Feb 9;7(2):e1000229. pmid:20161723
- 16. Gray RH, Wawer MJ, Serwadda D, Sewankambo N, Li C, Wabwire-Mangen F, et al. Population-based study of fertility in women with HIV-1 infection in Uganda. The Lancet. 1998 Jan 10;351(9096):98–103.
- 17. Ogilvie GS, Palepu A, Remple VP, Maan E, Heath K, MacDonald G, Christilaw J, Berkowitz J, Fisher WA, Burdge DR. Fertility intentions of women of reproductive age living with HIV in British Columbia, Canada.AIDS. 2007; 21
- 18. Agangi A, Thorne C, Newell ML. Increasing likelihood of further live births in HIV-infected women in recent years. BJOG: an international journal of obstetrics and gynaecology. 2005 Jul;112(7):881–8.
- 19. Blair JM, Hanson DL, Jones JL, Dworkin MS, Adult/Adolescent Spectrum of HIV Disease Project Group. Trends in pregnancy rates among women with human immunodeficiency virus. Obstetrics & Gynecology. 2004 Apr 1;103(4):663–8.
- 20. Schwartz SR, Rees H, Mehta S, Venter WD, Taha TE, Black V. High incidence of unplanned pregnancy after antiretroviral therapy initiation: findings from a prospective cohort study in South Africa. PloS one. 2012 Apr 27;7(4):e36039. pmid:22558319
- 21. van Benthem BH, de Vincenzi I, Delmas MC, Larsen C, van den Hoek A, Prins M, European Study on the Natural History of HIV Infection in Women. Pregnancies before and after HIV diagnosis in a European cohort of HIV-infected women. AIDS. 2000 Sep 29;14(14):2171–8. pmid:11061659
- 22. Sharma A, Feldman JG, Golub ET, Schmidt J, Silver S, Robison E, Minkoff H. Live birth patterns among human immunodeficiency virus-infected women before and after the availability of highly active antiretroviral therapy. American journal of obstetrics and gynecology. 2007 Jun 30;196(6):541–e1. pmid:17547887
- 23. Kabami J, Turyakira E, Biraro S, Bajunirwe F. Increasing incidence of pregnancy among women receiving HIV care and treatment at a large urban facility in western Uganda. Reproductive health. 2014 Dec 6;11(1):81.
- 24. Henshaw SK. Unintended pregnancy in the United States. Family planning perspectives. 1998 Jan 1:24–46. pmid:9494812
- 25.
Mohohlo M, Khosa T, Mohapi L, Gray G. Pregnancy related events in an Antiretroviral Treatment Program. InThe 2006 Implementers Meeting of the President’s Emergency Plan for AIDS Relief, Durban, South Africa. Abstract 2006 Jun (Vol. 113, pp. 13–18).
- 26. O'Shea MS, Rosenberg NE, Tang JH, Mukuzunga C, Kaliti S, Mwale M, Hosseinipour MC. Reproductive intentions and family planning practices of pregnant HIV-infected Malawian women on antiretroviral therapy. AIDS care. 2016 Aug 2;28(8):1027–34. pmid:26877194
- 27. Carter AJ, Bourgeois S, O'Brien N, Abelsohn K, Tharao W, Greene S, et al. Women-specific HIV/AIDS services: identifying and defining the components of holistic service delivery for women living with HIV/AIDS. Journal of the International AIDS Society. 2013;16(1).
- 28. Stephenson JM, Griffioen A. The effect of HIV diagnosis on reproductive experience. Aids. 1996 Dec 1;10(14):1683–7. pmid:8970689
- 29. De Vincenzi I, Jadand C, Couturier E, Brunet JB, Gallais H, Gastaut JA, et al. Pregnancy and contraception in a French cohort of HIV‐infected women. Aids. 1997 Mar 11;11(3):333–8. pmid:9147425
- 30. Fiore S, Heard I, Thorne C, Savasi V, Coll O, Malyuta R, et al, Newell ML. Reproductive experience of HIV-infected women living in Europe. Human Reproduction. 2008 Jun 20;23(9):2140–4. pmid:18567897
- 31. Massad LS, Springer G, Jacobson L, Watts H, Anastos K, Korn A, et al. Pregnancy rates and predictors of conception, miscarriage and abortion in US women with HIV. Aids. 2004 Jan 23;18(2):281–6. pmid:15075546
- 32. Wilcher R, Cates W. Reaching the underserved: family planning for women with HIV. Studies in Family Planning. 2010 Jun 1;41(2):125–8. pmid:21466112
- 33. Gipson JD, Koenig MA, Hindin MJ. The effects of unintended pregnancy on infant, child, and parental health: a review of the literature. Studies in family planning. 2008 Mar 1;39(1):18–38. pmid:18540521
- 34. Wilcher R, Cates W. Reproductive choices for women with HIV. Bulletin of the World Health Organization. 2009 Nov;87(11):833–9. pmid:20072768
- 35. Sanders LB. Women's voices: The lived experience of pregnancy and motherhood after diagnosis with HIV. Journal of the Association of Nurses in AIDS Care. 2008 Feb 29;19(1):47–57. pmid:18191768
- 36. van Benthem BH, de Vincenzi I, Delmas MC, Larsen C, van den Hoek A, Prins M, European Study on the Natural History of HIV Infection in Women. Pregnancies before and after HIV diagnosis in a European cohort of HIV-infected women. Aids. 2000 Sep 29;14(14):2171–8. pmid:11061659
- 37.
Public Health Agency of Canada, HIV/AIDS Epi Updates 2008, Editor. 2014, Public Health Agency of Canada: Ottawa, Ontario
- 38. Loutfy M, Greene S, Kennedy VL, Lewis J, Thomas-Pavanel J, Conway T, et al. Establishing the Canadian HIV Women’s Sexual and Reproductive Health Cohort Study (CHIWOS): operationalizing community-based research in a large national quantitative study. BMC medical research methodology. 2016 Aug 19;16(1):101. pmid:27543135
- 39. Israel BA, Schulz AJ, Parker EA, Becker AB. Review of community-based research: assessing partnership approaches to improve public health. Annual review of public health. 1998 May;19(1):173–202.
- 40.
Benoit C, Shumka L. Gendering the population health perspective: fundamental determinants of women’s health. Final report prepared for the Women’s Health Research Network. Vancouver, BC. 2007.
- 41.
Raphael D, editor. Social determinants of health: Canadian perspectives. Canadian Scholars’ Press; 2009.
- 42.
De Reus LA, Few AL, Blume LB. Multicultural and critical race feminisms. Sourcebook of family theory and research. 2005:447–68.
- 43.
Webster K, C.A., Proulx-Boucher K, Dubuc D, Nicholson V, Beaver K, Gasingirwa C, et al. Strategies for recruiting women living with HIV in community-based research: Lessons from Canada. Progress in Community Health Partnerships: Research, Education, and Action. In press (2017).
- 44. Abelsohn K, Benoit AC, Conway T, Cioppa L, Smith S, Kwaramba G, Lewis J, Nicholson V, O'Brien N, Carter A, Shurgold J. " Hear (ing) New Voices": Peer Reflections from Community-Based Survey Development with Women Living with HIV. Progress in community health partnerships: research, education, and action. 2015;9(4):561–9.
- 45.
Kaida A, C.A., Lemay J, O’Brien N, Greene S, Nicholson V, Thomas-Pavanel,. et al. in 23rd Annual Canadian Conference on HIV/AIDS Research 2014: St. John's, Newfoundland, Canada.
- 46. Phares TM, Morrow B, Lansky A, Barfield WD, Prince CB, Marchi KS, et al. Surveillance for disparities in maternal health-related behaviors—selected states, Pregnancy Risk Assessment Monitoring System (PRAMS), 2000–2001. MMWR Surveill Summ. 2004 Jul 2;53(4):1–3. pmid:15229409
- 47. Günthard HF, Saag MS, Benson CA, Del Rio C, Eron JJ, Gallant JE, et al. Antiretroviral drugs for treatment and prevention of HIV infection in adults: 2016 recommendations of the International Antiviral Society–USA panel. Jama. 2016 Jul 12;316(2):191–210. pmid:27404187
- 48. Wood E, Montaner JS, Chan K, Tyndall MW, Schechter MT, Bangsberg D, et al. Socioeconomic status, access to triple therapy, and survival from HIV-disease since 1996. Aids. 2002 Oct 18;16(15):2065–72. pmid:12370506
- 49. Hoffman IF, Martinson FE, Powers KA, Chilongozi DA, Msiska ED, Kachipapa EI, et al. The year-long effect of HIV-positive test results on pregnancy intentions, contraceptive use, and pregnancy incidence among Malawian women. JAIDS Journal of Acquired Immune Deficiency Syndromes. 2008 Apr 1;47(4):477–83. pmid:18209677
- 50. Bankole A, Keogh S, Akinyemi O, Dzekedzeke K, Awolude O, Adewole I. Differences in unintended pregnancy, contraceptive use and abortion by HIV status among women in Nigeria and Zambia. International perspectives on sexual and reproductive health. 2014 Mar 1;40(1):28–38. pmid:24733059
- 51.
Statistics Canada. Pregnancy Outcomes by age grou. Canada Vital Statistics, Editor. 2005, Statistics Canada
- 52. Elgalib A, Hegazi A, Samarawickrama A, Roedling S, Tariq S, Draeger E, et al. Pregnancy in HIV‐infected teenagers in London. HIV medicine. 2011 Feb 1;12(2):118–23. pmid:20807252
- 53. Sutton MY, Patel R, Frazier EL. Unplanned pregnancies among HIV-infected women in care—United States. JAIDS Journal of Acquired Immune Deficiency Syndromes. 2014 Mar 1;65(3):350–8. pmid:24189153
- 54. Ezugwu EC, Iyoke CA, Nkwo PO, Ezegwui HU, Akabueze JC, Agu PU. Unintended pregnancy among HIV‐positive pregnant women in Enugu, southeast Nigeria. International Journal of Gynecology & Obstetrics. 2016 Jan 1;132(1):60–3.
- 55. Haddad LB, Feldacker C, Jamieson DJ, Tweya H, Cwiak C, Chaweza T, et al. Pregnancy prevention and condom use practices among HIV-infected women on antiretroviral therapy seeking family planning in Lilongwe, Malawi. PloS one. 2015 Mar 26;10(3):e0121039. pmid:25811849
- 56. Schwartz S, Papworth E, Thiam-Niangoin M, Abo K, Drame F, Diouf D, et al. An urgent need for integration of family planning services into HIV care: the high burden of unplanned pregnancy, termination of pregnancy, and limited contraception use among female sex workers in Côte d'Ivoire. JAIDS Journal of Acquired Immune Deficiency Syndromes. 2015 Mar 1;68:S91–8. pmid:25723996
- 57. Westreich D, Maskew M, Rubel D, MacDonald P, Jaffray I, Majuba P. Incidence of pregnancy after initiation of antiretroviral therapy in South Africa: a retrospective clinical cohort analysis. Infectious diseases in obstetrics and gynecology. 2012 Jun 19;2012.
- 58. Burgos-Soto J, Balestre E, Minga A, Ajayi S, Sawadogo A, Zannou MD, et al. Incidence of pregnancy following antiretroviral therapy initiation and associated factors in eight West African countries. Journal of acquired immune deficiency syndromes (1999). 2014 Oct 1;67(2):e45.
- 59. Kaida A, Andia I, Maier M, Strathdee SA, Bangsberg DR, Spiegel J, et al. The potential impact of antiretroviral therapy on fertility in sub-Saharan Africa. Current HIV/AIDS Reports. 2006 Dec 1;3(4):187–94. pmid:17032579
- 60. Kaida A, Laher F, Strathdee SA, Money D, Janssen PA, Hogg RS, Gray G. Contraceptive use and method preference among women in Soweto, South Africa: the influence of expanding access to HIV care and treatment services. PloS one. 2010 Nov 5;5(11):e13868. pmid:21079770
- 61. Kaida A, Carter A, Ding E, Sereda P, Salters K, Webster K, et al. Dual contraceptive use among women living with HIV in Canada: Potential for women-centred HIV care delivery models. AIDS 2016. Durban, South Africa (poster)
- 62. Grossman D, Onono M, Newmann SJ, Blat C, Bukusi EA, Shade SB, Steinfeld RL, Cohen CR. Integration of family planning services into HIV care and treatment in Kenya: a cluster-randomized trial. Aids. 2013 Oct 1;27:S77–85. pmid:24088687
- 63. Harrington EK, Newmann SJ, Onono M, Schwartz KD, Bukusi EA, Cohen CR, Grossman D. Fertility intentions and interest in integrated family planning services among women living with HIV in Nyanza Province, Kenya: a qualitative study. Infectious diseases in obstetrics and gynecology. 2012 Jul 12;2012.
- 64. Oulman E, Kim TH, Yunis K, Tamim H. Prevalence and predictors of unintended pregnancy among women: an analysis of the Canadian Maternity Experiences Survey. BMC pregnancy and childbirth. 2015 Oct 13;15(1):260.
- 65. Floridia M, Tamburrini E, Tibaldi C, Anzidei G, Muggiasca ML, Meloni A, et al. Voluntary pregnancy termination among women with HIV in the HAART era (2002–2008): a case series from a national study. AIDS care. 2010 Jan 1;22(1):50–3. pmid:20390480
- 66. Joyce T, Kaestner R, Korenman S. On the validity of retrospective assessments of pregnancy intention. Demography. 2002 Feb 1;39(1):199–213. pmid:11852837