HIV status and treatment influence on fertility desires among women newly becoming eligible for antiretroviral therapy in western Kenya: insights from a qualitative study

As access to life-saving treatment for HIV with antiretroviral therapy becomes universally available across the world, it is important to understand how treatment for HIV may change HIV-infected women’s desires to have more children. We conducted interviews with HIV-infected women in Kenya who had previously never been on HIV treatment to help us understand what their desires for having more children are, and how the possibility of starting HIV treatment soon may change their desires. The results of our study suggest that HIV-positive status greatly affects these women’s desires for having more children, and mostly in a negative manner in discouraging them from having more children. The women thought that starting HIV treatment soon would improve their health status but not necessarily increase their desire to have more children. Instead, we found that a separate set of individual and societal factors comes into play in influencing their desires for having more children. Individual factors, such as their current age, the number of children they already have, and how many of those children are still alive or infected with HIV, more directly influence their desires for having more children. Societal factors, such as community norms for motherhood and health providers’ expectations of their desires, also play an equally important role in influencing their desires for having more children. The findings of this study are crucial for health care providers and HIV treatment programs to understand in order to meet their HIV-infected women’s reproductive health needs.

The latest World Health Organization (WHO) guidelines on HIV treatment now recommend initiating antiretroviral therapy (ART) in all individuals living with HIV, regardless of CD4 cell count or disease severity [1]. This recommendation is informed by a growing body of evidence that highlights the benefits of early ART initiation [2‐4], including reduced mortality [5] and morbidity among HIV-infected individuals [1, 6]. Expanding access to ART in resource-limited settings over the last two decades has transformed the disease to a manageable chronic illness [7, 8] and allows HIV-infected individuals an improved overall quality of life [9] in these settings. Additionally, a study conducted in Uganda demonstrates that HIV-infected individuals now have similar life expectancy to the general population [10].

With increased life expectancy and improved quality of life [5, 6, 11], HIV-infected individuals now face decisions and considerations that they previously did not encounter. One such consideration is their fertility desires [12]. Few studies have evaluated the impact of HIV infection on fertility desires and pregnancy rates among HIV-infected women, though some evidence supports lower fertility desires among HIV-infected women compared to their HIV-uninfected peers [13, 14]. However, this gap in fertility desires appears to diminish with improvements in the health status of HIV-infected women [15, 16]. Additionally, only one study has examined ART’s influence on fertility desires [17], and to our knowledge, none have examined the influence of earlier ART initiation at higher CD4 cell counts, where the anticipated benefit of improvement in health status is lower. Up until July 2014, the Kenyan HIV treatment guidelines recommended initiating ART only at CD4 cell counts <350 cells/mm³ or with WHO clinical disease stage of III or IV [18]. Thereafter, the Kenyan guidelines included ART initiation for individuals with CD4 cells counts between 350 and 500 cells/mm³ [19], and since July 2016, has recommended ART initiation for all HIV-infected individuals regardless of CD4 cell counts [20]. As access to ART becomes universal, it is imperative to understand the fertility desires among HIV-infected individuals so that HIV treatment programs can meet their patients’ reproductive health needs.

Given the importance of understanding factors that influence fertility desires among HIV-positive women and how access to earlier ART modulates their fertility desires, we sought to examine how earlier ART initiation may influence future fertility desires. We conducted this qualitative study among a group of HIV-infected but ART-naïve women of reproductive age from western Kenya who meet the new country guidelines for initiating ART.

We conducted in-depth interviews with 20 HIV-positive ART-naïve women between July and August 2014 attending one of 13 HIV facilities in western Kenya. These facilities are supported by Family AIDS Care & Education Services (FACES), a collaboration between the University of California, San Francisco and the Kenya Medical Research Institute [21]. These facilities were located in Homabay, Kisumu, and Migori counties and ranged from County hospitals to health centers; they have been described in greater detail elsewhere [22]. We purposively selected facilities in each county/sub-county based on its representativeness of the region, accessibility, and level of health service provision. At each facility, we used a convenience sample to recruit 1–3 women to interview.

Study staff approached the nurse in-charge at each facility at the beginning of the clinic day and requested in-person referrals of 1–3 women attending the facility that day who were 18–45 years of age, non-sterilized, HIV-positive, ART-naïve, and had a CD4 cell count between 350 and 500 cells/mm³ within the last six months. Therefore, we sampled women who were ART-naïve but would be initiating ART in the coming weeks under the new Kenyan guidelines. Since the policy change in CD4 count threshold for ART initiation had changed in Kenya the same month of our study initiation, study participants were not aware of this newly changed treatment policy at the time of conducting the interviews. The study staff approached the potential participants and assessed interest for study participation.

The interviews were conducted in DhoLuo by a trained, female interviewer and digitally recorded. The interviewer confirmed study eligibility, obtained written informed consent, collected basic demographic information, and then started the interview. The interviewer used a semi-structured interview guide to prompt discussions on the following themes: 1) definitions of family and children’s role in family/community; 2) personal, interpersonal, institutional, and societal factors influencing fertility desires; 23) influence of HIV-positive status on fertility desires; and 34) influence of future ART initiation on fertility desires. The interviewer transcribed the initial two interviews in DhoLuo and then translated these into English. Another member of the study staff verified the accuracy of the English translations against the audio file and DhoLuo transcripts and resolved any discrepancies. Thereafter, the interviewer translated the interview audio files directly into English. Transcripts were coded manually using a word processor, and the coded text was further organized into themes and subthemes in separate documents. Transcripts from the interviews were iteratively analyzed using inductive content analysis [23]. An initial codebook was developed from the interview guide, which was further refined with discussion and consensus as the initial transcripts were coded. The first three transcripts were double coded by two members of the study team (A.T. and R.C.P.) and differences in coding were resolved through discussion until consensus was reached. After all the data were coded, the investigators used an iterative process of reading transcripts, comparing and contrasting codes, and identifying convergent and divergent themes.

This qualitative study demonstrates a spectrum of beliefs on how HIV status and access to earlier ART may influence future fertility desires. The women indicated that their HIV positive status does influence—largely negatively—their fertility desires confirming the finding of other studies [15, 17, 24], though it was less clear if these desires would change after initiation of ART. Ultimately, the women indicated that children are highly valued by them and their communities; social norms, community expectations, and sociodemographic factors, such as their age and number of living children, appeared to have the strongest influence on their fertility desires.

That fertility desires differ by HIV status, and that HIV-infected women are discouraged from future pregnancies is not surprising. Maternal morbidity and mortality is a significant consideration for all women considering a pregnancy. HIV-infected women have to consider additional HIV-specific factors, such as the risk of perinatal or sexual transmission of HIV, their current health status, social expectations, health care provider’s recommendations, and the burden of raising a possibly infected child alongside caring for oneself and her surviving children [17, 25‐30].

While a positive HIV status has a negative influence on fertility desires, our data demonstrates it is less clear if fertility desires will change with earlier initiation of ART, while a woman is relatively healthy. Our analysis reveals that HIV-infected ART- naïve women are largely aware that initiation of ART facilitates improved health status and life expectancy, providing them the opportunity to lead a “normal” life and pursue their goals and desires. Yet, for some women, initiating ART and anticipating improved health status does not necessarily translate to increased fertility desires. A few cross sectional surveys have evaluated fertility intentions among HIV-infected women on or off ART, and found no association between ART use, regardless of duration of ART use, and desire to have future children [26, 31, 32]. However, none of these studies further distinguish fertility desires among women with higher CD4 counts and good health status; it is possible that actual, instead of anticipated, fertility intentions might change after ART initiation among these women with higher CD4 cell counts, for example, due to quicker recovery of health and normalization of life.

Ultimately, these women may have additional, stronger influences on their fertility desires than the anticipated quality of life improvements with ART initiation. For instance, age, parity, and educational status strongly influence fertility [17, 24]. Our data supports the findings that women who already have children are less likely to desire additional children, while younger women without any children express a strong desire to have children [14, 24, 33].

What is interesting, however, is how large of an influence community expectations and beliefs had on these women’s fertility desires. The women found themselves navigating opposing perceived community expectations of their fertility. The women made it clear that, on one hand, children are highly valued by them and their communities. Having children is closely linked with higher social status and anticipated future economic returns, placing a high value on parenthood, especially motherhood, a finding well-demonstrated in settings both within [25] and outside of the context of HIV [29, 34‐36]. On the other hand, these women perceived community expectations for HIV-infected women to not have future children, generally out of concern for future orphan-hood for existing children, and the community’s subsequent burden of care, or potential HIV transmission to infant or partner [17].

Nonetheless, our data highlights a few HIV-specific factors that may more strongly influence these women’s fertility desires than initiation of ART. Some women in our study considered raising a child an additional burden to caring for themselves, which was a priority for them. In addition, women with perinatally-infected children seemed less likely to desire additional children, largely due to the fear of having additional infected children [37]. Initiation of effective ART would, of course, lower the risk of transmitting future children; however, the women in our study did not seem to appreciate this future possibility. In general, these women should be better supported to realize their reproductive health goals, whether that is using safer conception options, including life-long ART, to conceive or effective contraception to prevent pregnancies [38]. Practically this requires HIV treatment programs to integrate reproductive health services into their mainstream HIV care and systematically and proactively inquire about HIV-infected individuals’ fertility desires, so that the appropriate services can be provided to those individuals.

Despite sampling from across a large region and saturation of themes, our study has several limitations. First, we only conducted one-time interviews which does not offer insight into changing fertility desires over time. Second, social desirability bias to underreport their fertility desires from fear of perceived negative approval from society [17] may have affected our study. To minimize such potential bias, the interviewer strived to ask questions in a balanced fashion and use non-judgmental body language. Finally, though we sampled women attending HIV facilities from diverse regions of western Kenya, there might be limits to the transferability of our findings.

Our findings suggest that initiating ART among relatively healthy HIV-infected women previously not eligible for ART may not necessarily change their fertility desires. While a few HIV-specific factors, such as perceived immune suppression from pregnancy or already having perinatally-infected children, may influence fertility desires, it is individual and societal factors, such as social norms and community expectations, that appear to be the major determinants of fertility desires among ART-naïve HIV-infected women. Ultimately, HIV-infected women should be better supported to realize their reproductive health goals, through integration of such services into HIV care, systematic inquiry of fertility desires, and lastly the use of safer conception options, including ART, or effective contraception for pregnancy prevention.