Introduction

Women’s right to control their fertility has been a recognized human right and an international public health priority for more than five decades. Family planning programs designed to provide women with information and contraceptive options have been heavily supported by donors throughout that period. When it became clear that mother-to-child transmission of HIV was an important source of new infections, the potential for enhanced access to reproductive health services and reducing mother to child HIV transmission led to promotion of family planning for women living with HIV [1,2,3,4].

To reify that integration, the World Health Organization and its partners in 2002 put forth a four-pronged strategy [5] which proposed interventions to prevent young women from acquiring HIV, to prevent women living with HIV from experiencing unintended pregnancies, to limit risk of mother-to-child transmission at the time of delivery, and to prevent acquisition of HIV by infants postpartum. That strategy guided HIV programs for more than a decade. Eventually, as it became clear that antiretroviral therapy (ART) could virtually eliminate risk of transmission of HIV from mother to child and also improved the health of people living with HIV, treatment for all became the dominant theme, decreasing the need for special emphasis on women of reproductive age to prevent vertical transmission of HIV [6]. The priority for family planning programs once again focused on helping women achieve their fertility goals.

In 2013, we published a systematic review of family planning for women living with HIV in low- and middle-income countries, covering studies from the years 1990 to 2011, which addressed family planning and HIV integration efforts principally designed to prevent vertical transmission of HIV [7]. We identified nine relevant articles, all from Africa, which showed that providing concerted information and support for family planning use, coupled with ready access to a wide range of contraceptive methods, was effective in increasing family planning use. However, effects on pregnancy overall were difficult to measure, and no studies assessed the effect of family planning on subsequent unintended pregnancy.

Integration of family planning into HIV services is now more commonly seen as an opportunity to provide women living with HIV with the ability to avoid unintended pregnancies. This development reflects increased treatment effectiveness, as well as growing emphasis on integrated health services more generally, to ensure a client-centered approach to care.

Here, we present an update of our systematic review with an expanded scope, recognizing that increased access and coverage of treatment for HIV has profoundly changed the context of interventions and programs. We assess the evidence for the impact of providing or promoting family planning services to women living with HIV on the outcomes of contraceptive uptake, intention to use contraception, and pregnancy incidence.

Methods

Objectives

This review is part of the Evidence Project, a series of systematic reviews of HIV behavioral interventions in low- and middle-income countries conducted jointly by the Medical University of South Carolina and the Johns Hopkins University Bloomberg School of Public Health. This review updates and expands our previous review on this topic published in 2013 [7]. We conducted this systematic review in accordance with the Preferred Reporting Items for Systematic review and Meta-Analyses (PRISMA) guidelines [8].

Eligibility Criteria

Studies were included in the review if they met the following criteria: (1) family planning counseling (not just information or education) was provided to women living with HIV who knew their HIV status; (2) the intervention was conducted in a low-income, lower-middle income, or upper-middle income country (as defined by the World Bank [9]); (3) the intervention was evaluated using a study design that compared post-intervention outcomes using either a pre/post or multi-arm study design (including post-only exposure analysis); (4) the article was published in a peer-reviewed journal from May 2010 (the end date of the previous search on this topic) through June 21, 2022.

Studies presented data only from women living with HIV, or from both women living with HIV and without HIV. For the second category to be included in the review, the article must have presented pre/post or multi-arm data separately for women living with HIV. No language restrictions were used; English translations were conducted when necessary. If two articles presented data for the same project and target population, the article with the longest follow-up was retained for analysis.

Search Strategy

We searched four electronic databases (PubMed, CINAHL, Sociological Abstracts and PsycINFO) following the search strategy presented in “Appendix”. Titles, abstracts, citation details, and descriptor terms were independently screened twice. Abstracts were used to screen for inclusion. Full-text articles were obtained which were then reviewed by two independent reviewers to select the final studies. Consensus was used to resolve any differences.

Study Selection

A study staff member initially screened studies based on titles and abstracts and excluded non-relevant citations. Two senior study staff then screened the remaining citations using inclusion criteria. The results of both screenings were merged and compared, and consensus was established by discussing any discrepancies. Final study selection was based on a thorough reading of the full text articles.

Data Extraction

Data were entered by two study staff members into a systematic coding form independently that included detailed questions on intervention, study design, methods, and outcomes. The two completed coding forms were compared, and discrepancies were resolved by consensus, with review and discussion with a senior team member as necessary.

Risk of Bias Assessment

Risk of bias of included articles was assessed using an eight-point scale [10] which includes the following items: (1) prospective cohort; (2) control or comparison group; (3) pre/post intervention data; (4) random assignment of participants to the intervention; (5) random selection of subjects for assessment, or assessment of all subjects who participated in the intervention; (6) follow-up rate of 80% or more; (7) comparison groups equivalent on socio-demographic measures; and (8) comparison groups equivalent at baseline on outcome measures.

Analysis

Due to differences in study design, type of intervention, and measured outcomes, we were unable to conduct meta-analysis. We therefore extracted key findings from each study and present them by intervention types offered or analyses conducted.

Results

Electronic database searching yielded 4960 potentially relevant citations (Fig. 1). An additional 12 possible articles were identified through other sources. Once duplicate records were removed, 3453 abstracts were reviewed by one screener and 3312 records were excluded as not meeting the inclusion criteria. The remaining 141 abstracts were then reviewed by two people who removed 60 abstracts. The full texts of the remaining 81 abstracts were reviewed by two people, of which 25 articles met the inclusion criteria [11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35]. However, two articles [13, 33] were excluded because they presented data that were duplicated and more fully presented in subsequent publications [12, 32], resulting in 23 articles included in the review.

Fig. 1
figure 1

Flowchart of the systematic search and review process

Full size image

All the included articles came from Asia or Africa. From Asia, there was one study each from Cambodia [30], India [19] and Thailand [22]. From Africa, there were six studies from Uganda [12, 13, 25, 26, 31, 34, 35], four from Kenya [15, 16, 20, 27], two from South Africa [18, 23] and one study each from Malawi [32], Botswana [17], Cameroon [21], Ethiopia [11], Nigeria [24], Swaziland [29], Tanzania [14] and Zimbabwe [28]. No comparative studies from other countries or regions were identified.

The assessment of risk of bias for included studies is presented in Table 1. Eight studies employed cross-sectional [11, 15, 26, 31, 35] or serial cross-sectional [14, 18, 25] study designs. Five studies used pre-/post designs [17, 21, 29, 30, 34]; two were retrospective cohort studies [32] and one used a time series design [22]. Four studies were randomized control trials [12, 23, 27] or cluster randomized trials [16], and three were nonrandomized control trials [19, 24, 28].

Table 1 Risk of bias assessment of articles included in systematic review
Full size table

The included studies fell into two general types. The first type implemented and tested a specific intervention, while the second type focused on evaluating policy/programmatic studies, including national policy, clinic-level strategies or integration more generally. The thirteen studies in the first group are presented in Table 2 which describes the nature of the intervention provided as well as its effect. The eight descriptive or evaluative studies in the second group are presented in Table 3. Two additional studies which focused on men are presented separately in the text.

Table 2 Studies presenting outcome data on current contraception use, unmet need for family planning (FP), intentions to use family planning and incident pregnancies following intervention
Full size table
Table 3 Studies presenting evaluative data on policy or programs to address current contraception use, unmet need for family planning (FP), intentions to use family planning and incident pregnancies without providing specific interventions
Full size table

Taken as a group, the included studies presented in both tables used a wide variety of outcome measures to assess outcomes, both at the individual level and at the macro level. Sixteen of the included studies measured some aspect of current contraception use [11,12,13,14,15,16, 19, 21, 23, 24, 27,28,29,30,31, 33], three measured unmet need for family planning [13, 24, 33], three captured some aspect of intentions to use family planning [14, 17, 20], and four measured incident pregnancies following an intervention or integration [15, 18, 19, 31]. The wordings and specifics of the measures used and the results obtained are included in Tables 2 and 3.

The interventions provided in these studies included in Table 2 can be grouped in three large categories. Nearly all of these studies provided intensified support for family planning use, including individual counseling. Only three studies did not mention this as an element of the intervention offered [13, 15, 19]. Five studies focused efforts on integration of family planning and HIV services, including on-site provision of family planning methods either in the HIV clinic or at a family planning clinic in the same complex [15, 16, 18, 19, 23]. One study tested a facilitated referral model [13].

Increases in use of modern contraception following an intervention were reported in four studies, of which three provided intensified support for family planning [12, 21, 24] and one provided facilitated referral [13] (Table 2). In Uganda, Atakunda et al. [12] reported a significant increase in continuous use of family planning in the enhanced care group receiving structured sustained family planning support compared to standard care (OR 1.75, 95% CI 1.24–2.95). Landolt et al. [21] found a significant increase in dual method use among sexually active adolescents in Thailand, as well as increasing use of long-acting reversable contraception up to 48 weeks after initial visit. Mudiope et al. [24] reported a significant 31.3% increase in uptake of family planning at 6-month follow-up among women receiving the intervention compared to controls (p < 0.001) but a subsequent decrease in the post-intervention period. Baumgartner [13], testing a facilitated referral model, reported a significant 12% increase (p = 0.013) in the proportion of sexually active women using a contraceptive method in Tanzania. In a study in India, Joshi et al. [18] found that sixty percent of integrated clinic attendees reached family planning services after referral compared to only 8% in the control arm.

One intervention study differentiated subsequent use of family planning by women’s fertility intentions. Asking women living with HIV if they desired to be pregnant in the following 12 months, Atukunda et al. [12] found low levels of desire for future pregnancies in the intervention group compared to the control group (OR 0.23, 95% CI 0.08–0.64, p = 0.002). Other studies addressed the issue less directly, either not reporting percentages of women who desired pregnancy [13] or identifying women “in need of family planning” without specifying if that was subjectively or objectively defined [21, 24]. Baumgartner [13] assessed the effectiveness of the facilitated referral intervention in reducing unmet need for family planning and found a small statistically insignificant 4% decrease.

Three of the intervention studies assessed subsequent pregnancy. In Uganda, Atukunda et al. found a significant decrease in pregnancy in the first year postpartum after intervention [1.9% in enhanced care versus 8.8% in standard care (OR 0.20, 95% CI 0.05–0.62)] [12]. The two other studies did not detect a difference in subsequent pregnancies following intervention [15, 19].

The studies that primarily described or evaluated aspects of integration (Table 3) reported on much the same outcomes as the studies that implemented and tested specific interventions. In Ethiopia, Aradom et al. [11] reported an adjusted odds ratio of 4.53 (95% CI 1.7–12.06) for use of modern contraception among women receiving counseling about family planning from their ART provider compared to those not receiving counseling. In a retrospective cohort analysis in Malawi, Tweya et al. [31] examined the effect of the integration of family planning services into a large ART clinic in 2011. Using electronic medical records, the authors found that contraceptive use increased from 28% in 2012 to 62% in 2016 (p < 0.001). Pregnancy rates decreased over the period, from 6.8 per 100 person-years in 2012 to 1.3 per 100 person-years in 2016 (p < 0.001). In a cross-sectional study comparing use of modern family planning methods by women attending a clinic with minimal family planning information compared to women attending a clinic with family planning integrated into HIV care in Uganda, Wanyenze et al. [33] found a small but significant difference: 57.9% of women in the integrated clinic used a modern method compared to 50% of women in the clinic with only basic information (p = 0.04). In Swaziland, Siveregi et al. [28] found that most women were using a modern contraceptive method, with use among women living with HIV (84%) higher than among women without HIV (72.3%). After counseling, use of long term or permanent methods increased from 15.3 to 42.4% among women living with HIV.

Only two studies included men. In a study nested within a cluster randomized trial in western Kenya, Onono et al. [26] assessed the impact of integrating FP and HIV services on women’s and men’s knowledge of and men’s attitudes toward family planning. In general, familiarity with family planning methods was relatively high among women at baseline and increased by endline. No difference was detected between the fully integrated versus nonintegrated services. No significant changes in knowledge scores for men were found over time nor by integration status of the clinic attended. In a study in Uganda, Vu et al. [32] measured the effect of peer support groups and peer-delivered interventions for youth living with HIV, both male and female, on a variety of key behaviors. After adjusting for gender, age, education, marital status, and relevant covariates in a multiple regression analysis, Vu et al. found significant increases in self-efficacy for condom and contraceptive use (adjusted odds ratio [AOR] 1.82, 95% CI 1.30–2.55), knowledge about HIV (AOR 1.83, 95% CI 1.29–2.61), condom use at last sex (AOR 1.72, 95% CI 1.18–2.51), disclosure of HIV serostatus to a sex partner (AOR 1.61, 95% CI 1.01–2.55) and use of modern family planning methods (AOR 1.7, 95% CI 1.1–2.7), among others.

Discussion

Most of the studies reviewed reported significant results that were moderately effective in increasing family planning use among women living with HIV in low- and middle-income countries. Not all interventions evaluated were effective, however. Given the range of contexts for these studies, the variety of interventions and the complex challenge they address, that finding is not surprising.

Overall, across a variety of research designs, integration of services was found to be effective either in increasing some measures of current use of modern contraception [11, 14,15,16, 19, 28, 30, 31] or in increasing intention to use modern contraception in the future [14, 20]. Providing contraceptive methods on site in integrated clinics was effective in increasing use of modern contraception in many studies, resulting in increases ranging from the modest to more than doubling the rate of the comparator [12, 15, 19, 27]. This was not the case in all studies. In South Africa, for example, an intervention failed to increase acceptance of an intrauterine device (IUD) provided on-site or of tubal ligation which required referral [17], and in Cambodia, the target population of sex workers remained steadfast in their use of condoms and did not adopt dual protection by accepting modern contraception [29]. Referral for family planning services was also effective in increasing contraception use in some but not all studies. In Tanzania, a facilitated referral model yielded a 12% increase in use of modern contraception [13], and a small study in Uganda found the referral model resulted in the majority of referred women accepting at least one effective family planning service offered [24]. In another study in Nigeria, referral efforts appeared to be largely ineffective [23]. In general, the immediate availability of family planning services on-site has the logical appeal of being one less barrier for women whose fertility intentions were clear to them. The empirical evidence reviewed here by and large supports that logic.

The promotion of family planning for women living with HIV has garnered considerable interest in recent years. Two recent systematic reviews have been published, one addressing the factors that facilitated or constrained the integration of family planning into HIV services [36], the other examining whether integration of FP and HIV services increases the uptake of contraception among all women regardless of HIV status [37]. Our current review evaluating the impact of such programs adds to that growing literature.

Originally intended to be an update of our previous systematic review [7], this review differs in important ways. The priority given to PMTCT, considered essential a decade ago and prominent in our previous review, has decreased as effective HIV treatment has become widely available in low- and middle-income countries. Our previous review focused on the impact of family planning counseling and services on key HIV risk behaviors, including the prevention of mother to child transmission. The studies in this review primarily focused on family planning and women’s contraceptive use to achieve their desired fertility goals. Specific HIV-related outcomes were less in evidence, especially in the later studies. The numerous family planning outcomes addressed came with increased varieties of measurement employed.

The range of interventions evaluated also increased beyond what we encountered previously. An emphasis on integration of FP services in HIV treatment settings was clear, either through direct offer of services or through referral. Some studies provided enhanced interventions as well, to motivate family planning acceptance and use among women living with HIV.

Our earlier review found only nine articles meeting similar inclusion criteria over a period of 21 years [7]. This updated review found 23 articles over 12 years, with more diversity in the countries where the research took place. Though involving men in reproductive health decisions and services has long been considered an important goal and remains so [36, 37], only two of these studies included data from men. We clearly may have missed studies that focused exclusively on men, as that was not an element of our inclusion criteria. However, the fact that so few of the papers involving women included both genders in recognition of the important role that men may play in contributing to, facilitating or impeding women’s decision to use family planning is noteworthy. This paucity of data on the role that intervening with men might have on enhancing uptake of family planning by women is a weakness of the extant literature.

The challenge faced by many of these studies was facilitating future intentions and decision- making by women faced with complex social, cultural, and health issues. Each of these studies endeavored to detect the effectiveness of interventions that tried, through various means, to provide information, motivation, counseling, support, and access to the means to prevent unintended pregnancies. Their success in increasing contraceptive use may be considered modest given the level of effort and attention provided to the women who participated. The complex intersection of fertility desires and intentions, access to needed support and methods, as well as perceived limits on women’s reproductive rights and on their agency once fertility intentions have been formed, all conspire to make the challenge that much greater.

Decisions about fertility are complicated and often not solely under the control of an individual woman. In the presence of HIV infection, which heightens the importance of well-informed decisions about future fertility, the complication increases, influenced by the health status of the woman and her partner, previous parity, her openness about her HIV status, expectations of longevity, and concerns or experiences of HIV-related stigma, to name but a few. That these interventions have helped women as much as they appear to have done is a notable accomplishment. Recognizing the many factors that impinge on women’s ability to fully exercise their reproductive rights is the focus of reproductive health justice [38]. The importance of an enabling environment that allows women to fully exercise their right to decide when, how, how often, with whom they have sex and with which outcomes is increasingly being recognized. The absence of such an enabling environment will limit women’s ability to act on even the best counseling and support they may receive through efforts like the ones included in this review. The limited role of men in these interventions, addressed only in two of these studies, underscores the important need to address their continued influence, either facilitating or impeding, in the development of such an enabling environment.

Conclusions

The development of integrated health services has become a goal within global public health. For women, integrating family planning and HIV services has garnered considerable attention. Originally driven by the need to prevent mother-to-child transmission of HIV, a key goal of HIV programs, that integration is now driven largely by the desire to help women achieve their fertility goals and protect their reproductive health. As such, family planning and HIV integrated services now are increasingly regarded as nearly equal partners, a laudable advance from the previous era when FP was largely a helper service to achieve an important HIV goal. In this review, we found that integrated FP and HIV services can help women living with HIV grapple with their complicated decisions about fertility and take the steps necessary to act on those.