Societal and individual drivers of fertility desires and intentions among people living with HIV: a cross sectional study of HIV clinic attendees in Soweto, South Africa

Background High proportion of people living with HIV (PLHIV) who are in the prime of their reproductive years desire to have children. There are limited studies that explore the range of fertility intentions for PLHIV. This study investigated the fertility desires and intentions of PLHIV and the associated factors. Methods This was a cross-sectional study of 442 PLHIV receiving antiretroviral treatment (ART) in health facilities in Soweto, an urban township that is situated in the City of Johannesburg in South Africa. STATA version 13 was used to analyze the data. Results The participants' mean age was 36.3 years, 70% were females, 79.6% had at least one biological child, and 36% had 3+ children. Almost half (47%) expressed the desire for children, saying that this was because they had no biological children, or their partners wanted children, or they wanted children of a particular sex, or were feeling healthy after taking ART. An increased fertility desire was associated with absence of biological children (AOR = 5.06, 95% CI: 2.11–12.1) and with being married (AOR = 2.63, 95% CI: 1.31–5.27). A decreased fertility desire was associated with being aged 36+ (AOR = 2.63, 95% CI: 1.31–5.27), having primary education (AOR = 0.11, 95% CI: 0.01–1.30) and having ≥4 years of ART duration (AOR = 0.45, 95% CI: 0.24–0.81). Conclusion Individual factors played a significant role in shaping the fertility desires of PLHIV in this setting. The high desire for children underscore the need to integrate reproductive health services in HIV and AIDS care and treatment services and develop safer conception programmes to help PLHIV to conceive and have children safely.


Introduction
Across Southern and sub-Saharan Africa (SSA), the vast majority of people living with HIV (PLHIV) are adults of reproductive age [1][2][3]. Just like the general population, PLHIV desire to have children. After learning of their HIV-status, they desire to start families or to have more children to achieve their desired family size [4][5][6][7][8]. In the past, policies in many countries discouraged PLHIV from having children in order to reduce vertical transmission. Consequently, being infected with HIV posed a huge threat to the expression of reproductive desires because of the potential risks of reinfections and the orphaning of existing and future children [8,9].
Medical achievements in the treatment and management of HIV/AIDS have seen a profound reduction in the level of risk posed for PLHIV by having children. The increased access to antiretroviral therapy (ART) and the availability of prevention of mother-to-child transmission (PMTCT) interventions have changed the patterns of morbidity and mortality globally. With HIV being viewed as a manageable chronic illness, fertility issues in PLHIV have altered in favor of childbearing through PMTCT and safe conception interventions [8]. ART has transformed the general desire to have children into an actual intention [2,10].
While a strong link has been established between improved health status and reproductive desire among PLHIV, fertility desire can be influenced by a number of factors that operate at the societal and individual levels [8,11]. At a societal level, socio-cultural expectations play a significant role in shaping the reproductive desires of PLHIV [6,[11][12][13][14][15]. For PLHIV, meeting family and societal obligations concerning reproduction may be more important than any risk of HIV transmission that may exist [16]. In SSA, research has documented the enormous pressure on women to have children, regardless of their HIV status [4,6,7,17,18]. Motherhood is a key component of a woman's identity and is important to her social status [14,16,19].
Equally, for men, not having children often results in stigmatization and loss of social status among their peers [20]. Fatherhood is critical, as their perceived masculinity and status in the community depends on their ability to have children to ensure the continuation of the family name and lineage [3,18]. In many societies in SSA the husband makes decisions regarding the couple's sexual activities, fertility, and use of contraceptives [11,21]. In these patriarchal societies, married women are often pressurized by their husbands to have children [22]. Not having a male child puts a couple under pressure to continue having children until a male heir is produced [23].
In South Africa the fertility desires of PLHIV, in particular of HIV-positive men, have not been sufficiently studied, yet it seems to be generally accepted that male partners have a strong influence on women's fertility-related desires and decisions [1,20,24]. The lack of data might be because of the healthcare seeking behavior of men. Data in South Africa show that health-seeking behavior of men with regards to HIV services is biased towards women, thus more women visit HIV clinics for ART [25]. The purpose of the study was to investigate the fertility desires and intentions of adult PLHIV and determined the associated factors. By including both males and females, the study will bridge the gap on the lack of data on male partners, albeit the number of males who access ART services is small. HIV and reproductive health programs should acknowledge the needs of PLHIV to have children and provide the necessary counseling and other services to support clients to conceive and have children safely [26,27].

Study design
A descriptive cross-sectional survey was conducted over a period of four months, from November 2017 to February 2018. The setting for the study was primary health facilities (PHC) in a health sub-district in Soweto, an urban township that is situated in the City of Johannesburg, in South Africa. Soweto is the biggest township in the country with a population of 1,695,047. The population is predominantly black (98.5%), males make up 49.6% of the population while females make up 50.4%. The sub-district where the study was conducted has 23 primary health facilities from which five were selected randomly for the study. All the facilities offer comprehensive primary health services including HIV testing and counseling, and the initiation and continuation of ART treatment for adults and children.

Study population
The population of the study consisted of PLHIV who were receiving ART at facilities in the health district. The participants were selected if they were between the ages of 21 and 49 years for women and 21 and 59 years for men, who visited the selected health facilities for their ART treatment during the period of the data collection. Those who were too ill to be interviewed were excluded from participating in the study. The sample size was determined using the Rao soft sample calculator [28], with a population size of 1500 at a confidence interval of 95% and a margin of error of 5% with a 50% response distribution rate. The sample size required was 369, and with a 20% adjustment to cater for potential refusals, 442 was the total sample size.

Data collection
A structured, piloted, interviewer-administered questionnaire was used to collect data. The questionnaire was developed in English and was translated into two local languages, namely IsiZulu and Sesotho, to cater for the diverse population of the township. The questionnaire was informed by extensive literature review and the objectives of the study [4,5,12]. It included demographic information, the period since the HIV diagnosis, the duration on ART, the disclosure of the HIV status to the sex partner, the perceived health status, the reproductive history, and the desire and intention to have children since the subject found out about his/her HIV status. The primary issue to be studied was fertility desire, which was defined as the desire to have children in future.
The participants were recruited in the mornings in the clinics as they came for their scheduled appointments for medication refill or check-ups, and interviews were conducted only after they had done with their business at the clinics. The investigators worked closely with the facility managers, who introduced them to the nurses responsible for HIV treatment and care. The nurses were introduced to the study by being given study information sheets and informed about the inclusion criteria. The facility staff particularly nurses identified patients who met the inclusion criteria and referred them to the researchers. Due to the sensitivity of the study, this recruitment method was appropriate to prevent accidental disclosure of the HIV status of the patients. Patients who met the inclusion criteria were subsequently referred to a room, which had been reserved for completion of the questionnaire in private. The purpose of the study was thoroughly explained to the potential participants, who were informed about voluntary participation, the confidentiality of their responses, and withdrawal from the study at any point without any penalties. Informed consent was then obtained from those who gave their consent to participate before the questionnaire was administered the lead author. There was general willingness to participate in the survey and none of the patients referred to the researcher refused to participate in the study. Although an about 70-100 people visit the different clinics HIV treatment daily, an average of 15-20 questionnaires could be completed per day.

Data analysis
Data were captured into Microsoft Excel spreadsheet before being exported to STATA version 13 for analysis. Descriptive statistics were computed to generate frequencies for all categorical variables, such as gender and marital status. Continuous variables such as age were summarised using means and standard deviation using a t-test. Bivariate analysis was performed between the dependent variable (the desire to have children) and each of the independent variables explored in the study, using Pearson's chi-square test of association. All variables identified to be significant at P < 0.05 were included in the multivariate model to determine the factors associated with fertility desire, using a backward stepwise logistic regression method. Odds ratios (OR) at 95% confidence intervals were obtained, as were p-values.

Ethics approval and consent to participate
The study was approved by Sefako Makgatho University Research Ethics Committee (protocol number: SMUREC/H/58/2017). Permission was granted by the Gauteng Department of Health-Johannesburg Health District (reference number GP_201733_699). All participants gave written informed consent before participation. To ensure anonymity, no personal identifiers were collected, from the participants

Description of the study participants
The description of the study participants are presented in Table 1. Of the 442 PLHIV who participated, 309 (70%) were female, their ages ranged from 22 to 57 years with a mean age of 36.3 years (SD ± 7.03 years). With regard to marital status, 20.8% were married, 29.4% were cohabiting with partners in a committed relationship, and 36.7% were single but having a steady sexual relationship. The majority (79.6%) had biological children, 9.7% were raising a child who was perinatally infected with HIV, and 6.6% had lost a child in infancy due to AIDS related illness. The majority (63.1%) had completed high school, 63.8% were employed.
A high proportion (43.4%) had known about their HIV status for four years or more, and 41% had received ART for four years or more. The level of disclosure was high, 86.2% had disclosed to at least one significant person, 69.3% had knowledge of their partners' HIV status. The majority (94.6) perceived their health status to be good.
Concerning fertility intentions, almost have (49%) desired to have a child, 52.2% intended to have one child, 46.4% were already trying for a baby, and over a third (36.8%) intended to have a baby in within the next three years.

Reasons for the fertility desire of the participants
Almost half (47%) of the participants desired to have children, a high proportion (39%) wanted to try for a child of a particular sex, 23% wanted to try for a boy child, and 11% wanted to try for a girl child, 29.7%) did not have children of their own, and 20% reported that their partners want more children. More women (n = 32) than men (n = 10) would consider child bearing because their partner want more children.
Of the 51% of participants who did not want children in future, the most commonly cited reason being having the desired number of children (42.5%). In addition, 21% indicated that they did not have a stable partner, whereas, 14% felt that they were too old to bear children.

Discussion
The study investigated the fertility desires and intentions of PLHIV on long-term ART in health facilities in Soweto, South Africa. The combined fertility desire was significantly high. We found that almost half (47%) of both the male and the female participants desired to have a child in the future. The combined rate of fertility desire of 47% was higher than those found in Uganda (31%), Tanzania (37%), Kenya (34%), and Malawi (33%) [6,8,9,26]. In contrast, the fertility rate in the current study was lower than those from Uganda (63.1%) and Nigeria (73.2%) [11]. The rate of desired fertility is generally high in SSA because of the context in which PLHIV live and the great value placed on parenthood [8,12,21]. The strong fertility desire observed in the current study has public health implications for HIV prevention. Mmbaga et al. [6] argue that strong fertility desire justifies the integration of HIV care and reproductive health so that appropriate services may be offered to PLHIV.
In the current study, the most common cited reasons for wanting children were the lack of existing biological children and a preference for a child of a particular sex. These findings are consistent with those of other studies [6,18,29,30]. Those who did not have children of their own were five times more likely to desire children than those who already had children. Other studies have made similar findings [6,18,[30][31][32]. The ideal number of children that individuals would like to have is a predictor of fertility desire.
Consistent with past studies in SSA [6,8,11,29], age was a predictor of fertility desire. We found that those older than 35 years had less desire for children than those below 36 years. Decreasing desire for children among older people was reported in several studies [6,9,22,33]. The majority of PLHIV in SSA are in the prime of their reproductive years, therefore, it is natural and to be expected that younger, unmarried PLHIV, would be eager to start families [6,11]. In addition, those who were married were 3 times more likely to have fertility desire than those that were not married. This is consistent with other studies where those in long term stable relationships considered child bearing [6,10,11,27,30]. Being married not only creates a sense of security but it provides a reliable support for the raising of children [27]. The PLHIV's desire for children underscores the need for counseling and reproductive services to help PLHIV to conceive and have children safely in the future [30].
Prior studies have found that PLHIV who have attained qualifications were more likely to have fertility desires [32,33] probably because they might be well-informed about the effects of ART on their health and are also more empowered to make informed decisions about their reproductive choices [32]. We did not find an association with higher qualifications, but we found that those who had primary level education were less likely to express fertility desires. Although the availability of ART has yielded good health outcomes for PLHIV and restored their reproductive desires [8,10,34], in this study, those who have been on ART for four years or longer were less likely to express fertility desire. The data showed that most (82%) of them were older than 36 years, an age group that did not desire children.
The results of this survey should be interpreted in the context of the following limitations. We did not collect clinical data such as viral load because of poor record keeping in most facilities, as a result, we could not assess the impact of viral load and CD4 count levels on the desire and intentions to have children, and instead we relied on the participants self-reported health perceptions. Although we wanted to assess the fertility desires of both men and women, the proportion of men compared to women in the facilities were we conducted the survey were small. Literature in South Africa shows that health-seeking behavior of men with regards to HIV services is biased towards women, thus more women visit HIV clinics for ART [25]. Lastly, our use of a convenience sampling technique to select the study participants affected the ability to generalise the findings beyond settings similar to that of the current study. Furthermore, we do not have any data on participants who might have refused participation because we relied on the nurse referring the potential participants to the researchers after consultation. Therefore, those who might have decline participation could have informed the nurse and were subsequently not referred to the researcher for completion of the questionnaire.

Conclusions
A desire to have children was expressed by almost half of our sample, supporting the view held that an HIV diagnosis does not eliminate a desire for fertility. The findings have implications for the development of guidelines and the design of programs that include discussions about future fertility in HIV counselling.
The study found that a fertility desire was driven mostly by individual factors. Increased fertility desire was associated with the lack of biological children and being married. Older age, low levels of education, longer duration on ART were associated with decreased fertility desire. Although not statistically significant, for female participants, the partner's fertility desire influenced their intentions to have children.
There is a need to integrate reproductive health services in HIV and AIDS care and treatment services and develop programmes to provide safer conception to help PLHIV to conceive and have children safely. While being male was not a predictor of a desire for children, men should be especially targeted by the interventions for safe conception. This underscores the need to design programs that are male-friendly to engage men in safer conception interventions.