Contraception values and preferences of people living with HIV: a systematic review

Abstract

Objective

Attention to the contraception values and preferences of those living with HIV is essential to meeting their reproductive rights and health needs. We systematically reviewed the literature on contraception values and preferences among women and men living with HIV.

Study design

We searched ten electronic databases for articles from 1 January 2005 through 27 July 2020 for qualitative and quantitative studies of the values and preferences for contraceptive methods among individuals living with HIV.

Results

Twenty-one studies, primarily from sub-Saharan Africa, met the inclusion criteria. Contraception values and preferences were shaped by several factors: availability, accessibility, and convenience; perceived effectiveness; safety and tolerability; dual protection; fertility desires; partnership dynamics; and provider recommendations. Male condoms were a frequently preferred contraceptive method, offering an affordable and accessible form of dual protection against HIV and unwanted pregnancy. Fears of infertility and side effects decreased interest in hormonal contraceptive methods. Financial burdens incurred by HIV management and a desire to reduce dual reproductive health and HIV care burdens influenced preferences. Healthcare providers contributed to contraceptive preferences of women living with HIV, informing perceptions of safety, tolerability, and effectiveness.

Conclusion

Contraception values and preferences among women living with HIV are complex and influenced by factors related and unrelated to their HIV status. Considering contraception values and preferences of people living with HIV will ensure that their autonomy and right to make decisions about the contraceptive methods best for them are upheld.

1. Introduction

Attention to the contraception values and preferences of people living with HIV is essential to meeting their reproductive rights and health needs. Those living with HIV have the right to make autonomous reproductive decisions, including the right to safely plan for the children they desire and to prevent unintended pregnancies. Sadly, paternalistic and proscriptive views and actions concerning the reproductive capacity of and appropriate contraception for people living with HIV, particularly women, has not been uncommon, with some women reporting being coerced into accepting undesired contraceptive methods [1]. Empowering women living with HIV and HIV-affected couples who do not desire pregnancy with a range of contraceptive options that allow them to make decisions that are right for them based on their fertility goals, values, and preferences is important to taking a rights-based approach to the reproductive health care of people living with HIV. Understanding values and preferences around contraception of individuals living with HIV is critical to informing programs that seek to address the reproductive rights and contraceptive needs of those living with HIV.

Contraception values and preferences are influenced by individual, social, structural, and technological factors and may change over time. Factors unique to HIV infection may affect fertility intentions, sexual behaviors, and intimate relationships, and form the basis of the values and preferences in relation to contraception among those living with HIV. For example, fertility intentions might decrease following an HIV diagnosis due to a variety of reasons, such fear of the effect of pregnancy HIV disease progression or fear of onward HIV transmission to a partner or child [2]. In these cases, women might prefer long-acting reversible or permanent contraceptive methods, such as hormonal implants or injections, or sterilization. Alternatively, fertility intentions might remain the same or even increase due to, for instance, social and cultural norms that strongly value childbearing and parenthood or to restore a sense of “normalcy” in response to HIV stigma [2–5]. Consequently, women might have stronger preferences for short-acting and easily reversible contraceptive methods, such as condoms or oral contraceptive pills (OCPs), which would make it easier to discontinue use in the future when they are ready to have more children. Those in HIV serodiscordant relationships might have stronger preferences for dual-method contraceptive options—the simultaneous use of a condom and an effective, non-barrier modern contraceptive (such as hormonal contraceptive or intrauterine device (IUD)), to provide protection against both pregnancy and HIV transmission or other sexually transmitted infections (STIs). In addition, HIV status disclosure to a partner in serodiscordant partnership may have implications for contraception preferences. For example, women living with HIV who have not disclosed their HIV status to a partner and/or fear violence from a partner might place a higher value on contraceptive methods that can be used covertly or that reduce HIV transmission to a partner [6].

Antiretroviral therapy (ART) may further affect fertility goals and sexual behaviors, and thus contraception values and preferences. Improvements in health and overall quality of life can contribute to increased sexual activity and improved immunological functioning from antiretroviral therapy can result in increased fecundity, which can shape the contraceptive needs and preferences of those living with HIV [7, 8]. Health improvements from ART might also contribute to increased fertility intentions due to renewed hopefulness about the future [8], and result in preferences for short-acting contraception options that can easily be discontinued when ready to attempt to conceive later.

Meeting the contraceptive needs of those living with HIV while considering their values and preferences not only ensures respect and dignity in reproductive health care, but also has important public health implications. Women living with HIV have high rates of unintended pregnancies [9]. They are also at greater risk of pregnancy complications and mortality during the pregnancy and postpartum periods than women without HIV [10–13]. HIV contributes to some 6% to 20% of all maternal deaths worldwide [12]. Though maternal mortality in women living with HIV has declined over the years as ART has become more widely available, the global burden of HIV-attributed maternal deaths remains unacceptably high [13, 14]. Preventing unintended pregnancies among women living with HIV by ensuring their access to a range of contraception options and supporting them to find the method that best matches their preferences are important to further reduce maternal mortality in high HIV prevalence settings and to eliminate pediatric HIV infection [15].

In this paper, we systematically review the literature on contraception values and preferences of people living with HIV.

2. Methods

This review was part of a wider set of reviews on values and preferences related to contraception globally [16]. This review was conducted according to both PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) [17] and ENTREQ (Enhancing Transparency in Reporting the Synthesis of Qualitative Research) guidelines [18]. We briefly describe the methods used to conduct the review below. A detailed description of the methods for the global review is available elsewhere [16].

2.1. Inclusion and exclusion criteria

Studies were included in the review if they presented primary data (quantitative or qualitative) on values, preferences, views, and concerns regarding contraceptive methods among women or men living with HIV. We defined values and preferences as the “collection of goals, expectations, predispositions, and beliefs that individuals have for certain decisions about their potential outcomes” [19], i.e., what matters when making decisions and how this shapes choices. We focused on contraceptive methods included in the 2015 WHO Medical Eligibility Criteria for Contraceptive Use (MEC) [20] and Selected Practice Recommendations for Contraceptive Use (SPR) [21] guidelines, which include not only modern contraceptive methods, but also fertility awareness methods, lactational amenorrhea methods, and coitus interruptus. We included studies published in peer-reviewed journals between 1 January 2005 and 27 July 2020 on the contraception values and preferences of women and men living with HIV in any country in the world. This timeframe was selected to include relatively recent studies, as values and preferences related to contraception may change over time. Studies had to present the results of quantitative or qualitative primary data collection. We imposed no further restrictions on study design. We excluded studies that only examined knowledge about contraceptive methods, prevalence of contraceptive use, contraceptive user characteristics, factors associated with contraceptive uptake or continuation, effectiveness/failure rates of different methods, or actual use of contraception as a proxy for preferences.

2.2. Search strategy

We searched 10 electronic databases to identify studies for inclusion. We also conducted secondary searching on reference lists of relevant articles, reviewed the reference lists of reviews on related topics, and asked the WHO MEC Guideline Development Committee for additional references that may have been missed. In addition to the secondary searching conducted for the global review [16], we conducted additional secondary searching on reference lists of relevant systematic reviews related to HIV and contraception [22–24]. A complete version of the search strategy in included in the global review methods paper [16].

2.3. Screening, data abstraction and analysis process

After removing duplicates, a single reviewer conducted an initial screening to remove irrelevant articles based on title, abstract, and key words. Two reviewers then independently screened the remaining citations, resolving discrepancies through discussion and consensus. Full-text articles were obtained for all citations deemed eligible, or possibly eligible, for inclusion. Included studies underwent data abstraction using standardized forms. We assessed study rigor for quantitative studies using an 8-item measure developed by the Evidence Project [25] and for qualitative studies using an adapted version of the Critical Appraisal Skills Programme (CASP) qualitative checklist [26]. We did not conduct meta-analysis because of the diversity of study designs and outcomes measured in the quantitative studies.

3. Results

The global review search identified 422 articles, of which 19 specifically mentioned people living with HIV. The secondary search of reference lists of systematic reviews related to HIV and contraception identified three articles not included in the global review search. Twenty-one out of these 22 articles met our inclusion criteria (Figure 1). One article was excluded because values and preferences of people living with HIV were reported in the context of achieving pregnancy safely, not pregnancy prevention. Table 1 presents descriptions of the 21 included studies. Nearly all studies were conducted in sub-Saharan Africa, specifically Uganda [27–31], Malawi [32–34], Kenya [35–39], Ethiopia [40, 41], South Africa [39, 42, 43], and Botswana [44]. One article presented findings from Cambodia [45] and two presented findings from Brazil [39, 46]. Of the included studies, 20 presented contraception values and preferences of women and five explored contraception values and preferences of men living with HIV [29, 30, 36, 46, 47], including as part of serodiscordant couples. A few articles presented findings of preferences among specific populations of women living with HIV, including female sex workers (FSWs) and people who inject drugs (PWID) [46], postpartum women [37], and adolescent girls [29].

Figure 1.

PRISMA flow diagram of screening and selection of studies on contraception values and preferences of people living with HIV.

Table 1.

Summary of characteristics and key findings of included studies on contraception values and preferences of people living with HIV.

Study ID	Study location and setting	Study population and sample size	Study design	Contraceptive method(s)	Outcomes assessed	Summary of key findings
Alene 2018	Amhara region, Ethiopia	Women living with HIV aged 15–49 years enrolled in HIV care with at least one ART clinic visit (N=803)	Cross-sectional survey	Modern contraceptive methods Fertility-awareness based methods	Preferences	The most used and preferred methods of contraception were: injectables (42.8%), male condoms (32.9%), and implants (11.1%). Although 42.2% of women reported accessing contraceptives at family planning clinics and 41.2% at ART clinics, two-thirds reporting preferring to obtain contraceptives at the hospital’s ART clinic.
Asfaw 2014	Addis Ababa, Ethiopia; Public health clinic	Women living with HIV aged 18–49 years receiving ART (N=1418)	Cross-sectional survey	Any modern contraceptive method	Preferences	Reasons for contraceptive use were protection against STIs (28%), prevention of unwanted pregnancy (21%), convenient to use (18%), easy to use (14%), does not interfere with sexual activity (8%), and advice of health workers (5%). Women on ART were more likely to report protection against STIs as reason for method choice (31.6% vs. 23.2% among ART-naïve women). ART-naïve women were more likely to report pregnancy prevention as reason (25.6% vs. 17.3% among women on ART).
Crede 2012	Greater Cape Town, South Africa; Public health clinics	Women living with HIV (n=265) and women without HIV (n=273) seeking child health services for infant 6 months or younger and who received antenatal care	Cross-sectional survey	Any modern contraceptive method	Preferences	Reasons for hormonal contraceptive method choice among women living with HIV were convenience (50.7%), health provider recommendation (48.9%), absence of side effects (8.3%), and effectiveness for preventing pregnancy (5.5%). Reasons for not using IUD among women living with HIV were prefer sterilization/already sterilized (25%), concern about insertion and removal procedure (20.5%), not sure if safe with current health status (20.5%), needing more information (11.4%) and being unsure (11.4%). Reasons for not using sterilization were: method permanence (52.6%), being too young/not married (31.6%), being afraid of the procedure (22.8%), and the surgical procedure (17.5%).
Haddad 2013	Lilongwe, Malawi; HIV care clinic	Women living with HIV aged 18–45 years on ART for at least 6 months seeking family planning services (N=200)	Randomized controlled trial	Copper IUD (n=99) DMPA (n=101)	Satisfaction Discontinuation rate Reason for discontinuation	Adherents of both methods reported high satisfaction. Twenty-nine women discontinued use of the copper IUD and 21 discontinued DMPA. Reasons for discontinuing copper IUD were side effects (38%), partner’s request (31%), desire to get pregnant (14%), and being no longer sexually active (7%). For DMPA, reasons for discontinuation were side effects (38%), no longer being sexually active (19%), partner’s request (14%), and desire to get pregnant (10%). Among those who discontinued, 96% and 58% were happy with the method and 94% and 81% would recommend the method to a friend at the visit prior to discontinuation and the discontinuation visit, respectively.
Haddad 2014	Lilongwe, Malawi; HIV care clinic	Women living with HIV aged 18–48 years screened for randomized controlled trial (N=281)	Cross-sectional survey within a randomized controlled trial	Combined contraceptive pills Progestogen-only pills Copper IUD DMPA Implants	Preferences Willingness to have IUD placed Reason for declining IUD placement	First-choice method was DMPA (58%), IUD (34.9%), and oral contraceptive pills (5.7%). Irrespective of first choice, 79% were willing to have an IUD placed at same visit. Of the 21% of women who declined IUD placement at the first visit, 37% were willing to have an IUD placed in the future. Reasons for declining IUD placement were desire for a different/specific method (39%), fear of pain, bleeding, infertility, or infection (25%), and having heard negative things about IUDs (25%).
Kakaire 2015	Kampala, Uganda; National referral hospital	Women living with HIV aged 18–45 years at least 4 weeks postpartum who desired to delay pregnancy for at least one year (N=703)	Randomized controlled trial	Copper IUD (n=349) Levonorgestrel (n=354)	Discontinuation rate Reasons for discontinuation	There were no differences in discontinuation rates between the copper and levonorgestrel IUD. Discontinuation rates at 1 year of placement was 8.6% in the copper group and 8.1% in the levonorgestrel group. Reasons for early discontinuation among copper IUD users included menstrual bleeding (21%), pain (21%), vaginal discharge (21%), partner complaint (21%), pelvic inflammatory disease (PID) (7%), and pregnancy (10%). Reasons for early discontinuation among levonorgestrel IUD users included menstrual bleeding (30%), pain (19%), vaginal discharge (19%), partner complaint (15%), pelvic inflammatory disease (11%), and pregnancy (7%).
Laher 2009	Soweto, South Africa Perinatal HIV Research Unit	Women living with HIV aged 15–40 years (N=42)	In-depth interviews (n=15) and focus groups (n=3 groups; 27 participants)	Any contraceptive method	Preferences Knowledge and attitudes	Several participants stated condoms were the ideal contraceptive method due to their lack of side effects and dual protection against unintended pregnancy and STIs. Perceived drawbacks of condoms were fear of breakage and necessity of male partner participation. Oral contraceptive pills were the least preferred method due to the burden of having to remember to take a daily pill. Consideration of potential interactions with ART affected women’s choice of contraception. Several participants reported being advised by health providers to switch from the pill to an injectable method because ART would render the pill ineffective. Two main side effects made injectables less desirable: vaginal wetness and amenorrhea. Vaginal wetness was reported as being viewed by male partners as undesirable. However, some women reported liking injectable methods because of convenience and secrecy.
Mayhew 2017	Central and Eastern provinces, Kenya; Family planning and postnatal care clinics	Women living with HIV aged 16–25 years accessing reproductive health services (N=288)	Cross-sectional survey (n=240) In-depth interviews (n=48)	Any modern contraceptive method	Preferences Satisfaction	Several women reported using condoms only after heavy promotion from healthcare providers but felt anxiety around condom failure and risk of unintended pregnancy. Several women expressed dissatisfaction with the pill (made them feel sick) or anxiety about forgetting to take the pill. Some were dissatisfied with injections, mainly due to heavy bleeding. The primary motivator for using long-acting reversible contraceptives (LARCs) or permanent contraceptive methods was effectiveness in preventing unwanted pregnancy. Women who preferred implants were motivated by lack of side effects and the long period of effectiveness, reducing required health facility visits. Several women wanted to use LARCs but were unsupported in their choice by partners or providers.
Mbonye 2012	Wakiso District, Central region, Uganda	HIV-positive women, pregnant woman, non-pregnant women, men, adolescent boys and girls, civic leaders, midwives, teachers. Focus groups targeted pregnant women aged 20 – 49 years, non-pregnant women aged 20 – 49 years, men aged 20 – 55 years and adolescents aged 15 – 20 years. Mean age for female survey respondents was 25.8 years (range: 14 – 49 years) (N=10895)	Mixed methods: cross-sectional household survey (n=10706), key informant interviews (n=66), focus groups (n=123)	Modern and traditional contraceptive methods (specifically male condom, OCPs, injections, fertility-awareness-based methods, withdrawal, and breastfeeding	Preferences Motivation for contraception use or non-use	Contraceptive use among women in the study was 50% among which 89% were on a modern method of contraception. Most popular modern method of contraception was injectable methods (53.2%) followed by pills (18.8%) and condoms (14.1%). Key informant interviews showed that most of them thought contraception was useful to them to prolong their lives through the prevention of HIV re-infection. Constraints for accessing contraception among women who are on ART included a fear of mixing ART drugs and contraceptives (47.8%). Other reasons for women on ART not using contraception included: Fear of side effects (24.6%); spousal disapproval (10.5%); wanting another child (11.9%); frequent stock-outs of contraceptives (5.2%).
Nattabi 2011	Northern Uganda; HIV care clinics	Women and men living with HIV aged 14–49 years attending outpatient HIV care clinics (N=502)	Cross-sectional survey (n=476) In-depth interviews (n=26)	Any contraceptive method	Preferences	Factors influencing low use of contraceptive methods included bad experiences with some methods, fear of side effects, health concerns, and in some cases partner opposition and religious affiliation (Catholicism). Some women preferred injectable contraceptives because they could use them covertly, without their husband’s knowledge.
Newmann 2013	HIV care facilities in Nyanza Province, Kenya	Healthcare providers with regular contact with patients in HIV care (N=31)	Cross-sectional survey Qualitative interviews	Any contraceptive method	Preferences of patients living with HIV Views on integrated HIV and family planning services	Healthcare providers reported concerns of their female patients living with HIV about side effects, such as irregular bleeding, from the use of contraceptive methods. They also reported that most women living with HIV tend to choose DMPA because they can use it covertly.
O’Shea 2015	Lilongwe, Malawi; Public hospital	Women living with HIV (n=210) and women without HIV (n=424) aged 18–45 years admitted to postpartum ward with live birth >28 weeks (N=634)	Cross-sectional survey	Any contraceptive method (particularly intrauterine contraception (IUC) and implants)	Preferences	Women living with HIV were more likely to plan to use condoms, sterilization, and less likely to use natural family planning, withdrawal, and oral contraceptives compared to women without HIV. Women living with HIV were more likely to report plans to use condoms and another modern contraceptive method compared to women without HIV postpartum (53% vs. 35%, p<0.001), and had lower preferences to having more children. There were no differences in interest in implant or IUC between women living with HIV and those without HIV postpartum. 64% of women living with HIV were interested in using the implant postpartum, while 21% were interested in IUC postpartum.
Patel 2014	Nyanza Province, Kenya HIV care clinic	Men living with HIV aged 27–55 years (N=21)	In-depth interviews	Any contraceptive method	Preferences	Men preferred obtaining contraceptive services at HIV clinics because of streamlined services, trust in HIV clinic providers, and feeling included in family planning discussions. Most men preferred condoms as a contraceptive method, mainly for their dual protection benefits.
Polis 2014	Rakai, Uganda; HIV care clinic	Women living with HIV aged 18–45 years interested in injectable contraception (N=357)	Randomized cross-over trial	DMPA-IM (intramuscular)  Sayana Press® (SP; subcutaneous DMPA)	Preferences Willingness to self-inject Satisfaction	At baseline the main perceived benefit of injectables was effectiveness. Though 91% of women had no concerns about using SP, only 68% had no concerns about using DMPA-IM, largely due to concerns about side effects with DMPA-IM. Most women predicting a preference for SP (57%) at baseline believed it would be less painful (38%) and cause fewer side effects than DMPA-IM (7%). Among women predicting a preference for DMPA-IM (41%), 57% were comfortable with using DMPA-IM, 21% believed it would be less painful, and 9% noted community experience with DMPA-IM. Most women (64%) preferred SP over DMPA-IM after experiencing both. Ever users who were very willing to self-inject at baseline had a 41% increased likelihood of selecting SP at 6 months. Over 90% of women chose to receive their injection in the arm.
Roxby 2016	Nairobi, Kenya	HIV sero-discordant couples (n=12 men; 12 women) and women living with HIV with incident pregnancy (n=10)	In-depth interviews (couples interviewed separately)	Dual methods (condoms and additional highly effective contraceptive method)	Preferences	Perceptions of side effects, male partner preferences, and reproductive desire were important factors in contraceptive preferences. Condom use was viewed as interfering with emotional aspects of sex and decreasing sexual pleasure. Dual contraceptive method use was viewed as redundant and a sign of possible infidelity. Those reporting dual contraceptive use reported concern that condoms might break and lead to an unintended pregnancy.
Schaan 2014	Gaborone, Botswana and surrounding villages of Mochudi and Moghoditshane; HIV care clinics	Women living with HIV aged 21–50 years on ART	Cross-sectional survey with 155 women living with HIV on ART	Any contraceptive method	Perceived effectiveness of method Perceived partner approval Perceived control over method use	Among women not desiring to have a child, 94% believed condom use was the most effective method to prevent pregnancy, 65% were concerned about hormonal contraceptive side effects, and 69% believed oral contraceptives could result in infertility. Women who planned to have a child believed condom use was the most effective method (84%), were concerned about side effects (54%), and believed oral contraceptives could result in infertility (62%). Some (19%) reported their partner did not approve of them taking hormonal contraceptives. Thirty-one percent of women planning to have a child and 45% of women not planning to have a child felt it was acceptable for a woman to take hormonal contraceptives and not tell her partner. Sixty-five percent reported most women have total control over having sex with a condom.
Shabiby 2015	Naivasha and Mbagathi district, Kenya; Hospitals	Postpartum women in postnatal ward (N=185: 91 women living with HIV, 94 women without HIV)	Cross-sectional survey	Implants	Preferences	Postpartum implant uptake was lower among women living with HIV (43%) compared to HIV-negative women (57%) (p<0.05). Reasons were having to think about it (30%), having to discuss with one’s partner (24%), desire to use another method (23%), and being afraid of side effects (12%). HIV-positive women were five times more likely to report fear of side effects and three times more likely to want to use another method. Women living with HIV were less likely to report needing to discuss with their partner as a reason for declining postpartum implant.
Telles Dias 2006	Six cities in Brazil: Belem, Salvador, Sao Jose do Rio Preto, Rio de Janeiro, Porto Alegre and Itajai	Vulnerable women who had used female condom for at least 4 months (n=225: 53 women living with HIV, 61 sex workers, 42 drug users, 31 partners of drug users; 68 other “at risk” women) and 29 men who were partners of women using female condoms (N=284)	In-depth interviews	Female condoms	Preferences Motivations for adoption	For women, the adoption of female condoms was motivated by: perceptions that the female condom is more durable and thus safer than male condoms; enhanced pleasure from stimulation by the external ring and less worry around condom breakage; additional comfort associated with lubrication; the ability to insert the female condom ahead of sexual intercourse so as not to disrupt the mood and to reduce the chance of forgetting to use a condom; and increased negotiating power and control over safer sex. Counseling sessions on using female condoms and negotiating their use were also viewed as facilitating their adoption. Men’s preferences for the female condom were influenced by the lowered burden on them to be responsible for putting on a condom.
Thyda 2015	Phnom Penh, Cambodia; HIV care clinic servicing most-at-risk women	Women living with HIV aged 18–45 years (N=250)	Pre-/post-integration cross-sectional surveys (250 pre-integration, 249 post-integration)	Any non-condom, modern contraceptive method	Preferences	Non-condom contraceptive use did not significantly change after implementation of integrated family planning services at the HIV clinic. Following integration, 34.5% of women surveyed believed IUD is not safe for women with HIV. Reasons for not using non-condom contraceptives following implementation were condom viewed as enough (79.5%), side effects (23%), health problems (18%), and being told people with HIV cannot use family planning methods (6.8%).
Todd 2011	Rio de Janeiro, Brazil Kericho, Kenya Soweto, South Africa	Women living with HIV aged 14–45 years attending health facilities, including HIV clinics (N=108; 36 per site)	Qualitative, free-list interviews	Any contraceptive method	Preferences	Factors that women living with HIV consider when selecting a contraceptive method reported by participants were: (1) prevention of HIV re-infection; (2) prevention of STIs; (3) effectiveness in preventing pregnancy; (4) prevention of HIV transmission to partner; (5) interaction with one’s own health/HIV disease progression; (6) partner agreement; (7) side effects; (8) interactions with ART; (9) abstinence. When asked about features of the ideal contraceptive method, participants reported prevention of pregnancy, STIs and HIV re-infection across all three sites. In Brazil, participants mentioned the need for the method to be controlled by women, to protect the partner from HIV or be undetectable by the partner. Contraceptives with few/no side effects and long-lasting effectiveness were reported as ideal across all three settings. Participants from South Africa were more likely than those in Kenya to state that an ideal method would maintain menstruation. Only participants from South Africa commonly reported an injection, no weight gain, and oral contraceptives as ideal features of a contraceptive method.
Wanyenze 2013	Kampala, Uganda; HIV care clinics (1 public clinic, 1 Catholic faith-based clinic)	Women, men, and adolescents living with HIV (n=98: 84 focus group participants, 14 interview participants) Healthcare providers (n=8)	Focus groups (84 adults and adolescents living with HIV) In-depth interviews with 14 women and men living with HIV, and 8 health care providers	Any contraceptive method	Preferences	Most adolescents and adults preferred condoms because of limited side effects, prevention of HIV re-infection or STI transmission, and that they are easily accessible, inexpensive, and easy to use. Some women cited challenges negotiating condom use and preferred methods they could use without their partner’s knowledge or permission (e.g., injectable, IUDs). However, they noted IUDs and implants were expensive and not easily accessible. Regarding oral contraceptives, women were concerned about taking additional pills. Adolescent girls were concerned pills and injectables would affect future fertility. Some women and men mentioned side effects with pills and injectables. However, some women who had used injectables said they did not experience major issues and preferred them since they did not need to use them daily but noted that the cost of injectables had increased. Some men knew of vasectomy but had mixed feelings and were not aware of anyone who had done it. Unlike adults, adolescents mentioned abstinence as a method of preventing pregnancy. Some wanted to use contraceptives but were unaware of the options and advantages or disadvantages of different types.

Eleven studies presented quantitative data [27, 28, 32–34, 37, 40–42, 44, 45], six presented qualitative data [29, 36, 39, 43, 46, 47], and four presented both quantitative and qualitative findings [30, 31, 35, 38]. Rigor assessments for each study are reported in Table 2 (quantitative studies) and Table 3 (qualitative studies). In general, studies were of moderate rigor. Quantitative studies seldomly randomly selected participants for assessment. Qualitative studies scored high across criteria except reflexivity (adequate consideration of the researcher-participant relationship).

Table 2.

Rigor assessment for quantitative studies (or studies with a quantitative component) on contraception values and preferences of people living with HIV included in the review.

Study ID	Cohort	Control or comparison group	Pre/post intervention data	Random assignment of participants to intervention	Random selection of participants for assessment	Follow-up rate >= 80%	Comparison groups equivalent at baseline in socio-demographics	Comparison groups equivalent at baseline in outcome measures
Alene 2018	No	No	No	NA	Yes	NA	NA	NA
Asfaw 2014	No	Yes	Yes	No	Yes	NR	NR	NR
Crede 2012	No	No	No	No	No	NA	NA	NA
Haddad 2013	Yes	Yes	No	Yes	No	Yes	NR	NA
Haddad 2014	No	No	No	NR	No	NR	NR	NR
Kakaire 2016	Yes	Yes	No	Yes	No	Yes	Yes	NA
Mayhew 2017	No	No	No	No	No	No	NA	NA
Mbonye 2012	No	Yes	No	Yes	No	NA	NA	NA
Nattabi 2011	No	No	No	NA	No	NA	NA	NA
Newmann 2013	No	No	No	NA	No	NA	NA	NA
O’Shea 2015	No	Yes	No	No	No	NR	NR	NR
Polis 2014	Yes	Yes	No	Yes	Yes	Yes	Yes	NR
Schaan 2014	No	Yes	No	NR	No	NR	No	NR
Shabiby 2015	No	No	No	No	No	NA	NA	NA
Thyda 2015	No	Yes	Yes	NA	NA	NA	Yes	Yes

NA: Not applicable; NR: Not reported

Table 3.

Rigor assessment for qualitative studies (or studies with a qualitative component) on contraception values and preferences of people living with HIV included in the review.

Study ID	Clear statement of research aims	Qualitative methodology appropriate	Research design appropriate	Recruitment strategy appropriate	Data collected in a way that addressed the research issue	Adequately considered researcher-participant relationship	Ethical issues considered	Data analysis sufficiently rigorous	Clear statement of findings
Laher 2009	Yes	Yes	Yes	Yes	Yes	No	Yes	Yes	Yes
Mayhew 2017	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Mbonye 2012	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Nattabi 2011	Yes	Yes	Yes	Yes	Yes	Unclear	Yes	Yes	Yes
Newmann 2013	Yes	Yes	Yes	Yes	Yes	Unclear	Yes	Yes	Yes
Patel 2014	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Roxby 2016	Yes	Yes	Yes	No	Yes	No	Yes	Yes	Yes
Telles Dias 2006	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Todd 2011	Yes	Yes	Yes	Yes	Yes	No	Yes	Yes	Yes
Wanyenze 2013	Yes	Yes	Yes	Yes	Yes	No	Yes	Yes	Yes

Studies discussed a range of contraceptive methods. Eleven studies reported on preferences concerning male condoms for contraceptive purposes among women and/or men living with HIV [29, 34–36, 39–41, 43, 44, 46, 47]. Over half of the 21 studies included in the review reported on hormonal contraception values and preferences among women living with HIV [28–31, 34, 35, 37–39, 41–44]. Several studies reported findings relevant to preferences for IUDs—either the copper or levonorgestrel IUDs—among women living with HIV. [27, 32–34]. Values and preferences concerning sterilization were reported in three studies from sub-Saharan Africa [33, 35, 42]. Two studies in Malawi [34] and Kenya [36] specifically reported on dual contraceptive use, or the use of condoms in addition to another modern contraceptive method. Only one study included in the review focused on female condoms [46]. Though included in the search strategy, we did not identify articles that reported specifically on other barrier methods, such as the diaphragm, cervical cap, sponge, or spermicides.

Below we present findings from the studies below by synthesized contraceptive values and preferences across method formulations.

3.1. Availability, accessibility, and convenience

Contraceptive availability, accessibility, and convenience featured prominently across studies as factors influencing method preferences among women living with HIV. Two studies in Uganda and Ethiopia specifically identified male condoms as a preferred contraceptive method due to their accessibility and convenience [29, 40]. The study from Uganda noted that adolescents living with HIV preferred male condoms for their ease of use and accessibility but reported challenges accessing information on other contraception options and were, thus, unaware of the advantages and disadvantages of the range of options available [29].

Convenience was also identified across five studies as affecting preferences for hormonal contraception [29, 30, 35, 41, 42]. Ugandan adolescents living with HIV stated cost and availability as preferences underpinning their interest in hormonal methods, specifically hormonal implants, injections, and IUDs [29]. In one study from Kenya, women living with HIV reported preferring hormonal implants because they offered long periods of effectiveness in preventing pregnancy without frequent hospital visits [35]. A cross-sectional survey conducted in Ethiopia with women enrolled in HIV care found that injectable contraceptives were preferred over male condoms and implants, and most women reported preferring to access contraceptives at the HIV care clinic, as opposed to a family planning clinic, due to service convenience [41].

3.2. Perceived effectiveness

Perceived method effectiveness shaped contraceptive preferences for women living with HIV. In a study conducted in Kenya in which most women living with HIV were married, had at least two living children, and had only up to a primary level education, women who did not intend to have more children preferred sterilization due to its effectiveness in preventing unintended pregnancies [35]. It should be noted that two of the four women in this study who chose to get sterilized had infants who had died of HIV/AIDS [35]. In the context of HIV care, concerns around reduced contraceptive effectiveness due to suboptimal adherence shaped women’s preferences for long-acting over short-acting methods, specifically OCPs. The additional pill burden of taking OCPs daily on top of ART deterred some women living with HIV from selecting OCPs as a preferred contraceptive method [29, 35], with some women expressing anxiety about forgetting to take the pill every day [35, 39, 43].

Stated preferences around perceived contraceptive effectiveness rendered barrier methods, specifically male condoms, undesirable for some people living with HIV. In a study from Kenya some participants reported choosing condoms as their method of contraception after heavy promotion from health providers but felt anxiety around condom failure [35]. Fear of male condom failure and the potential risk of unintended pregnancy resulted in some women and men living with HIV expressing lower preferences for male condoms [39, 43, 46, 47]. Only one study from Botswana identified male condoms as a preferred contraceptive method among women living with HIV due to its perceived effectiveness [44]. Another study from Brazil demonstrated that preferences for female condoms among women living with HIV were influenced by beliefs of their durability (and, therefore, effectiveness) relative to male condoms [46].

3.3. Safety and tolerability

Concerns about side effects was a common theme across studies, notably for hormonal methods [27, 29–33, 35–39, 42, 43]. In one study conducted in Botswana, 69% of the 155 surveyed women living with HIV on ART surveyed feared OCPs could result in infertility [44]. In a study from Uganda, adolescents living with HIV feared taking OCPs and hormonal injections because they perceived that these types of contraceptive methods could prevent them from having children in the future and reported having limited awareness of the advantages and disadvantages of various contraceptive methods [29]. A second study from Uganda found that women living with HIV perceived both OCPs and ART as “strong drugs”, and since they were already on ART there was concern that also taking OCPs might “make their bodies weaker” [31]. Concerns about side effects were particularly salient for both the copper IUD and levonorgestrel IUD due to concerns about insertion or removal, perceived increased risk of infection, and beliefs that IUDs were unsafe for women living with HIV [33, 39, 42, 45]. Fears of the surgical procedure also attenuated women’s interest in sterilization in one South African study [42].

In some cases, concerns about contraceptive tolerability stemmed from previous negative experiences with specific methods, particularly hormonal contraception. One mixed methods study in Uganda reported that side effects from past use of OCPs and hormonal injections shaped women’s preferences for non-hormonal methods [31]. In another study from Uganda assessing the acceptability of Sayana© Press (SP), a subcutaneous formulation of Depot medroxyprogesterone acetate (DMPA), versus intramuscular DMPA (DMPA-IM), 64% of women reported a preference for SP over DMPA-IM after having used both methods as part of the study [28]. Preferences for SP over DMPA-IM seem to have been influenced by fewer perceived side effects and pain.

Two studies conducted in Uganda and Malawi reported on women’s actual experiences with an IUD as enrollees in an RCT [27, 32]. Among these actual users, IUD discontinuation was primarily motivated by side effects, including menstrual bleeding, pain, and vaginal discharge.

3.4. Dual protection properties

Dual protection against HIV and other STIs as well as unintended pregnancy shaped preferences specifically for barrier methods (i.e., male condoms) in five studies [29, 39, 40, 43, 47]. In two studies, conducted in Malawi and Kenya, dual contraception (or use of condoms in addition to another contraceptive method) was reported as the preferred method of participants living with HIV over other methods [34, 36]. In Kenya, those women and men in serodiscordant relationships reporting a preference for dual contraception were concerned that the condom might break and felt using a second method was necessary to prevent an unintended pregnancy [36]. However, some women and men in the study also viewed dual contraception as redundant.

3.5. Fertility desires and intentions

Future fertility intentions also factored into contraceptive preferences and decision-making for people living with HIV. In a cross-sectional survey conducted in Malawi with 210 postpartum women living with HIV and 424 postpartum women without HIV, participants living with HIV were less likely to desire more children compared to women without HIV, and had higher preferences for condoms, and female or male sterilization [34]. The researchers found no differences in interest in implant or IUC between postpartum women living with and without HIV. Nearly two-thirds (64%) of women living with HIV were interested in using the implant postpartum, while 21% were interested in IUC postpartum [34]. A mixed methods study in Kenya also found fertility intentions influenced preferences for short-acting (i.e., condoms, OCPs) rather than permanent or long-acting (i.e., sterilization, implants, IUDs) modern contraceptive methods [35]. Fears of infertility deterred women living with HIV from specifically selecting IUD insertion in one study from Malawi [33]. Women living with HIV in South Africa were disincentivized from selecting more permanent methods of contraception, including sterilization, because of the method’s permanence in the context of future fertility desires [42].

3.6. Partnership considerations and dynamics

Partnership considerations weighed heavily on women’s contraceptive preferences and decision-making. In five studies conducted in Uganda, Malawi, Kenya, and South Africa, partner disapproval motivated refusal or discontinuation of a hormonal contraceptive method among women living with HIV [27, 30, 32, 35, 43]. In a study conducted in Uganda, both women and men living with HIV had negative perceptions of male sterilization through vasectomy, with some believing that it might result in male impotence [29].

For barrier methods specifically, the necessity of male partner participation for condom use also reduced preferences for male condoms [39, 43], particularly when there was a perceived risk of unintentional disclosure of one’s HIV status to a partner [39]. In one study from Brazil, preferences for female condoms among women living with HIV were motivated in part by increased negotiating power and control over safer sex, particularly when male condoms were unavailable or refused by a male partner [46]. Male partners of women using female condoms in the study expressed a preference for female condoms because it reduced the burden on them to put on a condom before sexual intercourse. Authors acknowledged the potential bias of men in favor of female condoms, since the male partners who agreed to participate in the study may have been more likely to have positive attitudes toward female condoms than male partners who refused to participate.

3.7. Provider recommendations

Healthcare provider recommendations shaped contraception preferences for some women living with HIV, informing their perceptions of safety, tolerability, and effectiveness [35, 40, 42, 43]. In some cases, provider recommendations limited women’s choices. In a qualitative study conducted in South Africa, some women reported being advised by healthcare providers to switch from OCPs to an injectable contraceptive method because ART would render contraceptive pills ineffective [43]. In another qualitative study conducted in Kenya, several women desired to use long-acting, reversible contraceptives, including hormonal methods, but felt unsupported in their choice by healthcare providers [35].

4. Discussion

4.1. Significance of findings

Data abstracted from 21 research articles, primarily from sub-Saharan Africa, highlighted many similarities in factors that influenced contraception values and preferences between women living with HIV and those without HIV. Values and preferences of women living with HIV that paralleled those of other women included convenience and affordability, perceived effectiveness, concerns around side effects, dual protection from STIs (including HIV) and unintended pregnancy, and partnership considerations. However, some factors unique to the experience of living with HIV shaped values and preferences.

Convenience emerged as a crucial dimension of contraception decision-making for women living with HIV. The women living with HIV in the studies included in this review expressed preferences for contraceptive methods requiring fewer routine visits to a health facility (e.g., condoms, injectables, IUD, implants vs. daily oral contraceptive pills), particularly given their clinic visit burden due to HIV care and treatment [29, 35, 40, 42]. The desire to reduce dual reproductive health and HIV care burdens (i.e., taking daily oral contraceptive pills and ART simultaneously) shaped preferences for non-daily contraceptive methods with fewer maintenance requirements. Some women living with HIV preferred to have contraception options accessible through HIV care clinics [41]. Attention to these issues can inform global efforts to scale up integration of HIV and contraceptive services so that women living with HIV receive better sexual and reproductive health care [48, 49].

Like women without HIV, women living with HIV expressed preferences for affordable contraceptive methods. Hormonal methods, specifically long-acting reversible contraceptives, were perceived as higher cost and, therefore, less desirable methods [29] compared to lower cost, widely available methods, like condoms [40], which offered comparable or acceptable levels of protection. For women living with HIV, additional financial burdens incurred by HIV management (e.g., transportation to clinics for ART, out-of-pocket care expenses) may heighten economic vulnerability and constrain contraceptive options. Increased market access to a range of subsidized, low-cost contraceptive methods can reduce financial barriers to contraceptive uptake and adherence for women living with HIV and their male partners.

Women living with HIV were reluctant to use certain types of contraceptives, particularly hormonal contraceptives, due to concerns of safety and tolerability, including fears of medical side effects [27, 29–33, 35–39, 42, 43] and infertility [33, 44]. In some cases, previous experiences with heavy bleeding and pain attributed to hormonal contraception [35] or an IUD [27, 32] shaped preferences. Though fears of side effects are also common for women without HIV [50], concerns around safety and tolerability of hormonal contraceptives for women living with HIV might be compounded by competing health issues associated with HIV. Acknowledging and making concerted efforts to understand the concerns expressed by women about the effect of hormonal contraceptives and IUDs on their bodies, while counseling them on potential side effects and alternative options for contraception, are critical to supporting women living with HIV make informed decisions about contraception.

Healthcare providers’ recommendations shaped contraception preferences of some women living with HIV, including their perceptions of safety and tolerability of different contraceptive methods [35, 40, 42, 43]. Some healthcare providers may not offer the full range of contraceptive methods, particularly to their patients living with HIV [38], despite WHO guidance that women living with HIV can generally use contraceptives, including hormonal contraceptives, without restriction [20]. There have been reports in the literature of women living with HIV having negative experiences when attempting to access contraception, including being told by healthcare providers that they could only use condoms or being coerced into accepting long-acting or permanent contraceptive methods [1]. Several WHO guidance tools can help healthcare providers facilitate conversations with clients to identify the best contraception options based on the client’s values and preferences, health status, and HIV drug regimen [20, 21, 48, 49, 51, 52]. These tools offer general recommendations to healthcare providers for ensuring quality of care in contraceptive services, including for women and adolescent girls living with HIV.

Though not unique to those living with HIV, partnership dynamics underpinned contraception values and preferences. Male partner preferences for contraceptive methods can reduce the autonomy of women to make voluntary and non-coercive choices around contraception. Women living with HIV, like those without HIV, experience difficulty in negotiating condom use [53]. Restricting their choice of contraceptives to those that rely on partner cooperation, i.e., condoms, may put women living with HIV at risk of unintended pregnancy and undermine their ability to exercise their sexual and reproductive right to determine whether and when to get pregnant [51]. Preferences for contraceptive options that can be used covertly have been expressed by women living with HIV and women without HIV [30, 54]. Women living with HIV, however, might also face the additional challenge of HIV status disclosure within their intimate relationships. Fears of unintentional HIV status disclosure to a partner can affect contraception preferences of women living with HIV [39].

4.2. Gaps in the literature

Despite our extensive search of the literature, we found only limited research on contraception values and preference for women and men living with HIV. Several gaps in the literature emerged from the review. Our search identified limited studies reporting on values and preferences around non-condom barrier methods (i.e., diaphragm, cervical cap, sponge, and spermicides) and no studies on non-modern contraceptive methods. Furthermore, we found limited data on values and preferences around sterilization. Specifically, none of the reviewed studies that reported on sterilization discussed the importance of childbearing as a sign of fertility and health, despite previous qualitative research findings that fertility preferences among women living with HIV are partly influenced by desires to be perceived as HIV-negative and healthy, and to regain social status following an HIV diagnosis, particularly in the context of high HIV stigma [3, 4, 55, 56].

We identified only five articles that included information on contraception preferences of men living with HIV [29, 30, 36, 46, 47]. Meaningfully involving not only women, but also men, living with HIV throughout the research, implementation and evaluation processes of sexual and reproductive health programs using participatory approaches will help to ensure that their values and preferences are considered throughout the process and to maximize uptake of contraceptive methods in HIV-affected couples that reflect their values and preferences [1].

Our review also only identified one article on contraception values and preferences of adolescents [29]. Globally, adolescent girls and young women make up a large proportion of new HIV infections and simultaneously are less likely to access the health services they need [57]. Quality health services, including contraception, are required to address the likely unique sexual and reproductive health needs of adolescent girls and young women living with HIV. The WHO has highlighted specific issues relevant to adolescent girls that are important to take into consideration when providing contraceptive services in the context of HIV treatment programs [48]. Specific considerations identified for adolescent girls include making accessible contraceptive methods that do not require daily regiments or regular resupply, which might be more convenient for adolescents, and providing information on dual method use since adolescents are disproportionately affected by STIs and allowing adolescents to weigh their options and make decisions that are best for them. Furthermore, since adolescents have been found to have higher discontinuation rates compared to adults [58–60], they may require greater support from health care providers in identifying the best contraceptive method for them based on their daily lives, values, fertility intentions, and preferences, including support in switching contraceptive methods if needed.

Healthcare providers play a key role in influencing contraception preferences of people living with HIV. However, there remains a gap in our understanding of how the quality of contraceptive services, including patient-provider communication on contraception, affects contraception values and preferences of people living with HIV. Additionally, more research on interventions that improve women’s self-efficacy and empowerment to make informed decisions around contraception and how healthcare providers can better support women living with HIV to meet their sexual and reproductive needs is essential [61].

Another critical gap was the lack of studies examining the perceptions of people living with HIV on drug-drug interactions between contraception, particularly hormonal contraception, and ART drugs, and how these perceptions may affect contraception values and preferences. There is some evidence that interactions between certain hormonal contraceptive methods and ART drugs could potentially reduce the effectiveness of contraception and increase the risk of unintended pregnancy [48]. Additionally, women and adolescent girls using certain TB regimens are advised against using combined oral contraception and other combined hormonal methods. There is consistent evidence that women living with HIV can use hormonal contraceptives without negative effects on HIV disease progression [62–64]. Our review identified a few studies that described participants’ fears that interactions between ART drugs and contraceptives, such as hormonal contraception, could result in harmful health effects, but few details surrounded these findings [31, 39, 43]. Understanding the perceptions of those living with HIV on potential drug-drug interactions between hormonal contraceptive methods and ART or TB drugs will be an important area for future research on contraception values and preferences.

4.3. Strengths and limitations

This review has several limitations. First, despite our thorough search process, we may have missed or inadvertently excluded some relevant articles. Second, nearly all studies included in the review were conducted in sub-Saharan Africa, limiting the generalizability of our findings since values and preferences are likely affected by culture, social norms, and other aspects of the local context. Third, many of the included studies did not report the formulations of contraceptives examined, which limited our ability to identify values and preferences for specific formulations of contraceptives, particularly hormonal contraceptives such as combined oral contraceptive pills vs. progestin-only pills. Despite these limitations, the review had key strengths. To our knowledge, this is the first review of values and preferences around contraceptive methods for women living with HIV. We employed a rigorous methodology for the review using well-established guidelines for systematic reviews, PRISMA [17] and ENTREQ [18]. Our comprehensive search of the literature had few restrictions, which allowed us to capture relevant data across diverse study designs, settings, and populations, including men who are often overlooked in contraception research.

Contraception values and preferences of women living with HIV bear many similarities to those of women without HIV. However, factors related to the experience of living with HIV might shape contraception values and preferences among those living with HIV, including dual contraception and HIV care burdens, financial burdens incurred due to HIV management, interactions between contraception and ART, and fears of unintentional HIV status disclosure to intimate partners. It is critical that women living with HIV are offered a full range of contraceptive options that will allow them to make informed decisions based on their own values and preferences.