Stigma and discrimination in healthcare settings

Stigma and discrimination manifest in broader society and in healthcare settings, and reflect levels and types of privilege, power, and disadvantage within society. Stigma and discrimination are related concepts with distinct differences [3]. Link and Phelan (2001) define stigma as the co-occurrence of “labeling, stereotyping, separation, status loss and discrimination” in contexts where power is exercised [4]. Discrimination, on the other hand, is defined as unfair and unjust actions towards an individual or group based on real or perceived status or attributes, medical conditions, socioeconomic status, gender, race, sexual identity, or age. Discrimination is a fundamental feature and expression of stigma, occurring both at the structural-level (societal conditions constraining opportunity or well-being) and individual-level (unequal or unfair treatment based on membership of a social group) [3]. Link and Phelan describe discrimination as the endpoint of the stigmatization process [4], while others view discrimination as a manifestation of the stigmatization process [5]. Both perspectives likewise place stigma and discrimination as fundamental causes of health inequities for three key reasons [3, 6, 7]: 1) stigma and discrimination influence several health outcomes through multiple pathways; 2) stigma and discrimination limit access to resources that can be used to avoid or minimize health risks or consequences; and 3) stigma and discrimination are related to health inequalities irrespective of time or place. Central to the conceptualization of stigma is the context and experience of power, privilege, and dominance that fosters environments of oppression and ‘othering’ of those who are stigmatized or discriminated against [8, 9]. When stigma or discrimination is experienced in healthcare settings, it is a violation of human rights [10, 11].

Stigma and discrimination are multi-level conditions that can differ across contexts, but there are common drivers, manifestations, and consequences present across settings and populations [12], including in healthcare settings. In Fig 1, we depict a multi-level stigma model that reflects these complexities within SRHR settings, mapped across the levels of internalized stigma, perceived stigma, enacted stigma, structural stigma, and layered stigma [13]. Internalized stigma refers to when a person with certain attributes is aware of public stigma about their attributes, agrees with these stereotypes, and applies the stigma to themselves [13]. Perceived stigma reflects an individual’s awareness of public stigma or beliefs that others hold stigmatizing thoughts about their condition or group [14, 15]. Enacted stigma refers to the manifestations of unfair treatment arising from adverse social judgment [16]. Structural stigma encompasses the societal conditions, sociocultural norms, and institutional policies that influence the opportunities and well-being of stigmatized groups [17].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Multi-level stigma model for sexual and reproductive health and rights.

This figure depicts the complexities of where stigma and discrimination exist within sexual and reproductive health and rights, and the levels where it can be confronted. Footnote: ⁱ Internalized stigma: person with certain attributes is aware of public stigma about their attributes, agrees with these stereotypes, and applies the stigma to themselves [13] ⁱⁱ Perceived stigma: individual’s awareness of public stigma or beliefs that others hold stigmatizing thoughts about a condition [14, 15] ⁱⁱⁱ Enacted stigma: manifestations of unfair treatment arising from adverse social judgment [16] ^iv Structural stigma: societal conditions, sociocultural norms, and institutional policies that influence the opportunities and well-being of stigmatized groups [17] ^v Layered stigma: interaction between multiple stigmatized identities within an individual or group [18].

https://doi.org/10.1371/journal.pgph.0000582.g001

Reproductive justice, a movement led by women of color in the United States, strives to achieve health equity, end discrimination and oppression, and challenge medical hierarchies through the recognition that the intersectionality of social, political and economic identities shape peoples’ abilities to access safe, appropriate, and respectful SRHR services [19, 20]. Through the lens of reproductive justice, we can better understand the intersecting influences of laws and social policies on people’s and communities’ health and well-being, and how these laws and policies contribute to longstanding injustices and structural oppression. Tantamount to achieving reproductive justice is acknowledging, then dismantling, oppressive systems that prohibit all people from achieving their fullest potential. Reproductive justice, therefore, can extend conceptualizations of stigma and discrimination from the interpersonal level, to better implicate the institutions and systems that perpetuate stigma and discrimination at the structural level. Similarly, global evidence has shown that women are mistreated during childbirth, which includes both structural and interpersonal discrimination while seeking care [21]. Eliminating stigma and discrimination within healthcare settings is critical to the attainment of reproductive justice and respectful care for all.

Structural and individual experiences or consequences of stigma and discrimination within sexual and reproductive healthcare have been well documented. Institutional and broader health system and social policies may drive stigma and discrimination in healthcare settings [12], where manifestations of stigma are both overt and covert [12, 21, 22]. Likewise, health workers’ perceptions, fears, belief systems, negative attitudes, moral distress, and lack of awareness about stigma, of a health condition, or the population may contribute to stigma in SRHR settings [12]. Individuals who are disadvantaged by systems of power and seeking sexual and reproductive health services may receive unfair or unequal treatment or have worse health outcomes compared to those who are in more privileged positions. Research has consistently shown that individuals who experience stigma and discrimination in healthcare settings may delay or forego seeking healthcare in the future [23]. Once individuals from stigmatized groups access the health system, they may be denied care or experience delays in receiving care. They may also be provided lower-quality care, receive care that is not culturally appropriate, or experience mistreatment such as discrimination, verbal and physical abuse, and denial of care [12, 24, 25]. Collectively, this can result in dissatisfaction and loss of trust in the health system, resulting in a vicious cycle of delaying care or not seeking care at all [26]. Additionally, it can lead to delayed diagnoses and initiation of care, as well as lower adherence and engagement in care, leading to poor health outcomes [27, 28]. The experience of discrimination itself has profound impacts on people’s physical and mental health as well as their general wellbeing [29–31].

Layered stigma refers to the interaction between multiple stigmatized identities within an individual or group [18]. For example, in the United States, persistent disparities exist in maternal mortality where Black women are three times more likely to die than non-Hispanic white women, and these experiences occur within a history of racist reproductive policies [32]. These reproductive policies intentionally practiced non-consensual or involuntary sterilization on Black women as a way of limiting the population of the perceived inferior group [33], and are a clear representation of structural racism. In Australia, racist government policies forcibly removed babies and children from Aboriginal families as part of the white Australia ‘assimilation policy’, leading to profound intergenerational trauma, loss of identity, and grief [34]. During apartheid South Africa, the Afrikaner government racialized and weaponized family planning by providing tax incentives for white women to procreate, while simultaneously promoting contraception for Black and Coloured women to limit fertility [35]. Present-day disparities are likewise driven by racism, which manifests in various ways including Black and Indigenous women lacking access to sexual and reproductive services, abuse in health care settings, assaults on migrant women’s reproductive autonomy, including forced sterilizations in detention centers, and the ongoing stress of detrimental colonial processes and living in race-conscious societies [36–40].

Stigma and discrimination also drive inequities in sexual and reproductive outcomes based on people’s social status, including socioeconomic status, marital status, and age. Prior studies have shown women of low socioeconomic status (measured variably by household wealth, caste, education, literacy, and employment status) are more likely to have poorer experiences during childbirth than those of higher socioeconomic status [41–43]. A World Health Organization (WHO) study on the mistreatment of women during childbirth in Ghana, Guinea, Myanmar and, Nigeria showed that adolescents and younger women were twice as likely to be physically abused and four times as likely to be verbally abused during childbirth, compared to older women, and 3% of women reported experiencing stigma or discrimination [22, 44]. Other forms of stigmatization based on gender and sexual orientation, disability, disease conditions such as HIV, STIs, tuberculosis, leprosy, substance use, and mental illness, as well as the type of care such as infertility and abortion services have been extensively documented as a barrier to care [24, 26, 41, 45–47]. Furthermore, the use of certain health services can be stigmatizing in and of themselves (such as abortion, STI, and HIV services), and thus can result in stigma and discrimination of healthcare users, providers, and communities.

Strategies to reduce stigma and discrimination in healthcare settings

Given the multi-level drivers and manifestations of stigma and discrimination, sustainable and scalable strategies to reduce stigma and discrimination in sexual and reproductive healthcare settings likely need to reflect this complexity. To date, much of the literature on strategies to reduce stigma and discrimination have been in the HIV field, with comparably less research on sexual and reproductive health. Nyblade and colleagues (2019) synthesized evidence on facility-based interventions to reduce stigma in HIV, mental illness, and substance abuse and identified six main approaches to reduce stigma broadly classified in Box 1 [12]. Many interventions identified had multiple components to reduce stigma [12], and the intervention mechanisms of action can be summarized as follows: improving healthcare provider awareness of stigma and contact with the stigmatized group may translate into improved practices of empathy in clinical encounters. Training healthcare providers on concrete tools and approaches to address stigma and work with stigmatized groups transforms abstract theories of stigma into concrete action in how they can provide better care. Policy reform may create more enabling environments for safe, respectful, inclusive, and anti-racist care without creating further barriers to certain groups seeking care. Lastly, strengths-based and community-driven approaches can empower individuals and groups who experience social, cultural, and economic oppression to achieve their SRHR by drawing on their own strengths and assets to lead initiatives or programs to tackle stigma and discrimination.

While perceptions, experiences and impacts of stigma and discrimination in health have been increasingly well-documented, a critical gap remains in understanding the evidence on strategies to reduce stigma and discrimination in sexual and reproductive healthcare. We aim to address this gap by conducting a systematic review of strategies (inclusive of interventions, programs, and policies) to reduce stigma and discrimination in sexual and reproductive healthcare settings, in order to describe types and characteristics of strategies, assess effectiveness, and synthesize key stakeholder experience and perceptions of the strategies.

Types of studies

For the quantitative component, randomized and non-randomized trials, pre-post studies (with or without a control group), interrupted time series, and other designs that compare the strategy to reduce discrimination with usual care were eligible for inclusion. Our operational definition of what constitutes a ‘strategy’ was inclusive of approaches such as policy reform, introducing new models of care, healthcare provider training (workshops, sensitization training, simulation), mystery clients, and community-based approaches targeting healthcare users. Studies published in abstract form were not eligible for inclusion, unless additional information could be obtained from study authors. Where we identified study protocols, we forward-reference searched to identify any results publications for relevance. We included studies that focused on strategies with the following characteristics: 1) studies that directly aimed to reduce stigma or discrimination in SRHR, or reducing inequity or promoting equity by reducing stigma and discrimination in SRHR; 2) studies that included a quantitative outcome related to SRHR and stigma or discrimination (e.g. experiences of stigma or discrimination, provider perceptions of stigma and discrimination); and 3) strategies that targeted healthcare users or communities, providers, health facilities or systems, health laws, or policies.

For the qualitative component, primary studies that used qualitative or mixed-methods designs to evaluate user or provider experiences of strategies to reduce stigma or discrimination were eligible for inclusion (e.g. process evaluations, ethnographies, case studies, phenomenological studies). Studies were eligible for inclusion regardless of whether they were conducted alongside studies of effectiveness (sibling studies) and did not need to have a comparison group to be eligible.

Topics of interest, types of participants, and settings

We included quantitative, qualitative, and mixed-methods studies that focused on strategies to reduce stigma and discrimination in sexual and reproductive healthcare settings. We have defined sexual and reproductive healthcare settings for this review as the following services, based on the Guttmacher–Lancet Commission on Sexual and Reproductive Health and Rights [1]:

• Maternal health services, including preconception care (pregnancy and infertility testing, counselling, and services), antenatal care (including STI screening and treatment, and preventing mother-to-child transmission of HIV (PMTCT)), childbirth care, and postpartum mother and baby care up to six weeks after birth (including immunization, breastfeeding)
• Contraceptive counselling and services, including STI screening
• Safe abortion and post-abortion services
• Reproductive tract cancers including cervical cancer counselling and services
• Gender-based violence screening
• Infertility testing and treatment

We excluded STI and HIV counseling, testing, and treatment services if conducted outside the context of contraceptive or maternal health services, as these topics have been well documented in the literature [12, 27, 28]. We excluded ‘comprehensive sexuality education’ as this is typically delivered in school-based settings, and the setting of interest in this review is healthcare settings. We adopted an intersectional approach to understanding discrimination by including studies that explore discrimination based on (but not limited to) race, ethnicity, Indigenous identity, social status, gender, sexuality, dis(ability), age, religion, migration or visa status, and the intersections between these identities.

Studies that included perspectives of healthcare users, family members of healthcare users, community members, healthcare providers, policy-makers, or other key stakeholders were eligible for inclusion, with no restrictions to sociodemographic characteristics or identity. Studies conducted in any country globally, and in any type of setting where sexual and reproductive healthcare is received (e.g. health facilities, community-based care, home-based care, or other types of institutional-based care) were eligible.

Search methods and study selection

We searched the following databases from inception to date of search, without any limits on language or publication date (S2 Text): MEDLINE (April 6, 2022), CINAHL (March 29, 2022), and Global Health Ovid (April 13, 2022). In addition to database searching, we reviewed the reference lists of included studies, conducted a forward citation search of all included studies using Web of Science, and conducted a grey literature search using OpenGrey (www.opengrey.eu). We collated all citations identified from different searches into Covidence (Covidence systematic review software, Veritas Health Innovation, Melbourne, Australia, www.covidence.org) and removed duplicates. Two review authors independently assessed each record for potential eligibility, and excluded references that did not meet the eligibility criteria (MAB, PAA, MVC, OJO, ANW). Two independent reviewers assessed full texts of potentially eligible studies, and disagreements were resolved through discussion and consensus with a third reviewer (MAB, PAA, MVC, OJO, ANW). For title and abstract assessment of studies published in languages that none of the review team are fluent in (languages other than English, Spanish, or French), we carried out initial translation through Google Translate. If the translation indicated agreement with the inclusion criteria, or if the translation is insufficient to decide, we consulted other colleagues in our networks to assist in assessing full text for inclusion.

Data extraction and critical appraisal

We designed a data extraction form for this review to extract data on study setting, sample characteristics, objectives, guiding frameworks, study design, strategy design and components, data collection tools and analysis methods, and author conclusions. For quantitative studies, we extracted data on the outcomes of interest related to discrimination (e.g. percentages, odds ratios, relative risks, prevalence estimates). For qualitative studies, we extracted the qualitative findings including author themes and participant quotations. Two independent reviewers critically appraised the methodological limitations of included studies using the Cochrane ROBINS-I tool [50] for quantitative studies and an adaptation of the Critical Appraisal Skills Programme (CASP) tool (www.casp-ul.net) for qualitative studies (MAB, PAA, MVC). Disagreements were resolved through discussion or involving a third review author where necessary.

Data management, analysis and synthesis

We used a narrative synthesis approach, which is a particularly useful approach when analyzing data from different types of studies [51]. A narrative synthesis approach focuses on the interpretive synthesis of the narrative findings of the research to sort and analyze studies into more homogenous groups based on critical components such as study design, study setting, program components, types of participants, type of health service and health topic, types of outcome, and direction or magnitude of effect [51]. There are four key elements of narrative synthesis recommended for implementation-focused narrative reviews, and these elements are analyzed iteratively throughout the review and synthesis process [51]. First, we developed a preliminary synthesis, which consisted of an initial textual descriptive analysis of included intervention studies and their findings. This preliminary synthesis allowed us to identify and evaluate initial factors, components, and processes that affected the construction of analytic outputs from the subsequent analysis steps. We used this textual descriptive analysis to group and cluster extracted data based on similar features, such as type of strategy, study location, and context, whether the strategy was designed with or by the stigmatized group, and type of participants (Table 1). Second, we explored the relationships between and across studies to understand how critical study design and intervention design factors may have influenced the likelihood of implementation success. Key relationships of interest included the relationships between study designs, levels of engagement, and magnitude and directionality of key findings (Table 2). Third, we used a thematic analysis approach [52] to synthesize the qualitative evidence of key stakeholder perspectives and experiences of the strategies. This included line-by-line coding of findings from primary studies and organization into descriptive and analytic themes and interpretations [53]. Fourth, we developed a model to map the strategy aims and outcome measurement across the different levels where stigma can be confronted, reflected in Fig 3. We assessed the robustness of the synthesis using multiple iterative methods to reflect on the methodological quality of the primary studies included in the synthesis and the trustworthiness of our analysis [51]. As described above, we assessed the methodological limitations of included studies using different tools appropriate for different study designs, and considered throughout the analysis process how to minimize bias, for example by ensuring that studies of equal technical quality are given equal weighting, and by clearly stating eligibility criteria across each step of the review (S2 and S3 Tables).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Characteristics of included studies.

https://doi.org/10.1371/journal.pgph.0000582.t001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Quantitative measurement approach outcome evaluation (among n = 9 studies with quantitative outcomes).

https://doi.org/10.1371/journal.pgph.0000582.t002

Throughout all stages of this review, we practiced critical reflexivity both as individuals and as a review team, which contributed to improving the robustness of the synthesis. This enabled us to consider, acknowledge and reflect on how our own lived experiences, employment, training, perspectives on discrimination and sexual and reproductive health services, and other factors shaped and influenced how we designed and conducted the review, synthesized, and interpreted the findings. We noted at the start of this review that the strength of our team comes from our own diversity: we have professional expertise and experience in sexual and reproductive health, public health, social sciences, medicine, obstetrics, epidemiology, social and reproductive justice, global health, and First Nations health. We are from different racial, ethnic, and religious backgrounds, and some of us are first- or second-generation migrants. We currently work at academic institutions, and at organizations that provide patient and community care in high-resource settings, and our projects regularly engage with people who are disadvantaged by existing systems of power at both global and local levels.

Description of included studies

Table 1 describes the characteristics of included studies. Half of the included articles (6/12) were conducted in the United States [54–59], one study conducted in each of Bangladesh [60], Canada [61], South Africa [62], United Kingdom [63], and Zambia [64], and one multi-country study conducted in three sub-Saharan African and three Latin American countries (countries not specified) [65]. Despite no date limits on the search, all included articles were published between 2003 and 2020. The topics within sexual and reproductive health varied substantially in the included articles: five were on abortion [54–57, 65], two on sexual health and HIV [60, 62], and one each on pregnant women with substance abuse [58], PMTCT [64], sexual health [59], sexual health, substance abuse, and STIs and blood-borne infections [61], and genitourinary medicine, family planning, and women’s health [63]. Most articles (8/12) collected data from participants who were health and/or social workers [54–56, 58, 59, 61, 62, 65]; three articles included both health workers and healthcare users [60, 63, 64], and one article included only healthcare users (women who had abortions) [57]. Six of 12 included articles described only quantitative results [55, 58, 60, 61, 64, 65], three were qualitative evaluations of strategies [54, 56, 57], and three were mixed-methods evaluations of strategies [59, 62, 63]. Among the articles that included quantitative measures, one was a cluster randomized controlled trial [64], and the remaining articles used quasi-experimental designs, including six articles using a pre-test, post-test single group design [55, 59, 60, 62, 63, 65], and two articles used a post-test only design [58, 61]. We used the ROBINS-I tool to assess the risk of bias in the quantitative studies (S2 Table outlines detailed critical appraisal). Most (8/9) studies had an overall critical risk of bias, primarily due to bias from confounding (typically due to no control group or adjustment for potential confounders), selection of participants (typically due to self-selection into the intervention group), bias due to missing data (substantial loss to follow-up or blank surveys) and measurement of outcomes (typically because participants who knew they were in the intervention group reported the impact of the intervention on their knowledge or behavior). One study had a serious risk of bias due to deviations from intended interventions, missing data, and measurement of outcomes. We used an adaptation of the CASP tool to assess methodological limitations in qualitative studies (S3 Table). These studies had minor to moderate concerns, due to limitations in data collection, reflexivity, and data analysis.

Narrative synthesis

Quantitative evaluation of impact.

Fig 3 describes the quantitative measurement approaches and outcome evaluation among the nine articles with quantitative outcomes. All articles measured health worker-reported outcomes [54, 55, 58–61, 63–65]. Three articles measured both health worker-reported and client-reported outcomes [60, 63, 64]. Four articles used validated measurement tools [55, 60, 64, 65]. One study also reported retention in care at 12 months for babies with mothers living with HIV [64].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Characteristics of included interventions.

This figure presents the characteristics of included interventions, mapped to the six main strategies to reduce stigma identified by Nyblade and colleagues [45].

https://doi.org/10.1371/journal.pgph.0000582.g003

Given the heterogeneity in outcome measurement (e.g. different tools, scales and measurement approaches) and evaluation (e.g. reporting percentages, means, odds ratios, and beta-coefficients), meta-analysis was not possible. Fig 3 narratively reports the impact of the interventions on the relevant outcomes of interest. In general, strategies that gave providers information about health conditions and/or stigma among the population group of interest reported increased self-reported awareness of the challenges faced by and decreased assumptions made about the population group. However, there were limitations in whether newfound knowledge was translated into changed practice or improved confidence in caring for population groups of interest, as these outcomes were typically not measured or reported by either providers or users. Of the three articles that evaluated user-experiences after the intervention [60, 63, 64], there were mixed results on the impact of the strategy on their experiences of care. In client exit interviews, Geibel and colleagues found that healthcare users were more likely to discuss being a member of a key population group and disclosing sexual activity to providers, and less likely to report the provider acted in a discriminatory way [60]. Phiri and colleagues found that women living with HIV were more likely to report social support and reduced perceived stigma from healthcare workers, despite no effect on retention of their babies in care at six or twelve months [64]. Kinn and colleagues found that service integration increased perceived confidentiality and happiness with overall care, but had no effect on the quality or availability of services [63]. Of note, the three articles that evaluated user experiences after the intervention collected data at two and six months after intervention [64], six months after the intervention [63], and six and twelve months after the intervention [60], which limits the evaluation of the sustainability of the intervention effects over time.

Mapping interventions and outcome measurement to the levels at which stigma exists and can be confronted.

Fig 4 maps the nine studies with relevant quantitative outcomes on stigma or discrimination to the different levels at which stigma can exist and be confronted. First, we mapped the studies based on which level or levels of stigma (internalized, perceived, enacted, structural, layered) the strategies aimed to address and how they were measured (Tables 1 and 2). Where strategies aimed to address or measure across more than one level of stigma, we indicated as such. Then we visually mapped the strategy aims and measurement approaches across the different levels of stigma to explore where the aims and measurement approaches were similar and different.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 4. Mapping included intervention aims and outcome measures using a multi-level stigma model.

This figure depicts a mapping of the nine studies with relevant quantitative outcomes on stigma or discrimination to the different levels at which stigma can exist and be confronted. First, we mapped the studies based on which level or levels of stigma (internalized, perceived, enacted, structural, layered) the strategies aimed to address and how they were measured (Tables 1 and 2). Where strategies aimed to address or measure across more than one level of stigma, we indicated as such. Then we visually mapped the strategy aims and measurement approaches across the different levels of stigma to explore where the aims and measurement approaches were similar and different.

https://doi.org/10.1371/journal.pgph.0000582.g004

Four articles (4/9) aimed to intervene and measured within the same level of stigma [55, 60, 62, 65]: Martin et al. 2014 and Mosley et al. 2020 both aimed to address and measured internalized stigma among abortion providers; Duby et al. 2019 aimed to address and measure perceived stigma among health workers and service users (sexual health); and Geibel et al. 2017 aimed to address and measure perceived and enacted stigma among health workers and service users (youth SRHR).

Five articles (5/9) had differences in the levels of stigma that the strategy aimed to intervene on, and the measurement approaches used [58, 59, 61, 63, 64]. These articles all aimed to address higher levels of stigma (structural and/or enacted stigma), but measured more proximal levels of stigma. For instance, Kinn 2003 aimed to address structural stigma by reconfiguring health services and measured the impact of service reconfiguration on perceived and enacted stigma [63]. Similarly, Jadwin-Cakmak aimed to address structural, enacted, and perceived stigma in sexual healthcare for gender- and sexuality-diverse youth after training workshops, and measured the impact on enacted and perceived stigma [59].

Strengths and limitations

Our review has several limitations. We excluded strategies that aimed to increase access to health services (such as policy-level abortion reform, and increasing access to facility-based birth), unless stigma or discrimination was an outcome measure. While this may have excluded some structural interventions that could potentially address discrimination related to societal conditions that constrain opportunity or well-being, we note that increasing access alone does not necessarily equate to a reduction in stigma or discrimination. We also excluded strategies that did not aim to reduce stigma or discrimination, but may have unintentionally reduced stigma or discrimination, for example, strategies to improve person-centered care. Given the complexity of addressing stigma, we believe that effective strategies should explicitly aim to reduce stigma and discrimination and measure the impact of the strategy on stigma and discrimination-related outcomes. Our review also has several notable strengths. Finally, we explored how we could assess the differences in intervention effects based on the “doses” of the intervention (once-off, or repeat). However, we found that due to the heterogeneity in outcomes reported, and multiple (and often many) outcomes reported in the quantitative studies, we are unable to conclude anything meaningful about the impact of a once-off versus repeat training on outcomes of interest. We hypothesize that given the complexity of addressing stigma and discrimination, a once-off training would be less likely to evoke change (especially sustainable change), and that measurement of this in the included studies is limited by evaluation of intervention impact typically occurring close to the time of the intervention. This may be particularly true for interventions targeting areas other than increasing provider awareness, which we hypothesize would be likely to need repeat and sustained engagement with stakeholders to influence systems change.

We used a systematic review approach, which increases the credibility, reliability, and transparency of the findings. Our broad inclusion criteria for types of strategies and types of studies (quantitative, qualitative, and mixed-methods) also ensured we captured the range of studies relevant to the topic. Our international team of researchers has broad and complementary experience across public health, clinical practice, and social and reproductive justice that enriched the quality and scope of the review.

Implications for policy and practice

The findings from this review have important implications for policy and practice. Individuals who experience stigma and discrimination when accessing sexual and reproductive health care are at risk of harm and poor health outcomes. But beyond that, it is a violation of their human rights. Thus, strategies are urgently needed to eliminate stigma and discrimination in SRHR settings. In designing, implementing, monitoring, and evaluating interventions to reduce stigma and discrimination, it is essential that the views of the group or community experiencing stigma and discrimination are incorporated. Ideally, such interventions should be led or co-led by representatives from the particular group or community [78], to ensure that their needs and experiences are addressed. We note that while the inspiration for this review stemmed from our work on mistreatment during childbirth and respectful care, we did not identify any strategies to promote respectful care or reduce mistreatment during childbirth that explicitly included strategies to eliminate stigma and discrimination. Moving forward, we encourage researchers, programmers, and policy-makers to ensure that this critical component of respectful care is not neglected.

Health care providers need to be supported to undertake appropriate education and training to understand and overcome discriminatory care practices. However, training health care providers in and of itself is not enough to drive change [8]. Health services need to enact structural and policy reform to create safe, inclusive, and respectful environments for people both providing and accessing SRH services. In some instances, stigma and discrimination can affect both the healthcare users and providers of sexual and reproductive health care, as can occur with abortion care [79]. As such, it is likely that effective interventions will be complex, operating across multiple levels to maximize the likelihood for change.

Interventions that address stigma and discrimination across all layers, especially at the enacted and structural levels are lacking. It is at this societal and political level where enabling legal and policy contexts and social norms can strongly influence the care provided and received by people accessing SRH services. Efforts to reduce legal restrictions and drive broader social and cultural change can have an impact at an individual, community, and institution level. Activists can also play a major role in not only encouraging community awareness, but also policy change. These broader societal, cultural, and legal shifts can have a significant influence on health care providers, influencing decision making, referral processes, and care provided.