Stigma in health facilities: why it matters and how we can change it

Stigma is a powerful social process that is characterized by labeling, stereotyping, and separation, leading to status loss and discrimination, all occurring in the context of power [1]. Discrimination, as defined by the Joint United Nations Programme on HIV/AIDS (UNAIDS), is the unfair and unjust action towards an individual or group on the basis of real or perceived status or attributes, a medical condition (e.g., HIV), socioeconomic status, gender, race, sexual identity, or age [2]. It has also been described as the endpoint of the stigmatization process [1]. Stigma is brought to bear on individuals or groups both for health (e.g., disease-specific) and non-health (e.g., poverty, gender identity, sexual orientation, migrant status) differences, whether real or perceived.

Health condition-related stigma is stigma related to living with a specific disease or health condition. Such stigma may be experienced in all spheres of life; however, stigma in health facilities is particularly egregious, negatively affecting people seeking health services at a time when they are at their most vulnerable. In health facilities, the manifestations of stigma are widely documented, ranging from outright denial of care, provision of sub-standard care, physical and verbal abuse, to more subtle forms, such as making certain people wait longer or passing their care off to junior colleagues [3‐6]. As a result, stigma is a barrier to care for people seeking services for disease prevention, treatment of acute or chronic conditions, or support to maintain a healthy quality of life [7‐19]. Within the health system, stigma towards a person living with a specific disease undermines access to diagnosis, treatment, and successful health outcomes [8, 20‐28]. Stigma also impacts the well-being of the health workforce because healthcare workers may also be living with stigmatized conditions. They may conceal their own health status from colleagues and be reluctant to access and engage in care [4, 29‐31]. Yet, stigma reduction is not a routine part of the way in which health services are delivered or evaluated, nor is it regularly integrated into pre-service and in-service training of all cadres of healthcare workers. This correspondence article explores how stigma is currently being addressed in health facilities across medical conditions, discusses gaps arising from a scan of the literature, and the potential for synergies across disease stigmas that could be harnessed for a joint response to more than one disease stigma. Specifically, for a variety of health conditions, we aimed to examine the health condition stigma addressed; intervention target populations, delivery, approaches, and methods; stigma drivers targeted; and evaluation methods and quality.

While recognizing that stigma is context-dependent, health condition stigmas in health facilities also display common features across countries and conditions in terms of certain stigma drivers, manifestations, and consequences [32‐38]. This is particularly the case with stigma drivers, or factors considered to produce or cause stigma [3]. Within health facilities, common drivers can include negative attitudes, fear, beliefs, lack of awareness about both the condition itself and stigma, inability to clinically manage the condition, and institutionalized procedures or practices [3, 32, 35, 39‐43]. Healthcare workers may fear infection, the behaviors of the stigmatized group (such as drug use or erratic or unpredictable actions), or mortality associated with the condition [3, 20, 32, 33, 35, 39, 40]. They may also experience moral distress based on their personal disapproval of behaviors associated with diseases, which may lead to stigmatizing reactions that impair their abilities to be effective providers, undermining quality of care [3, 20]. Healthcare workers may be unaware of how stigma manifests and affects people, and may therefore not be cognizant of the stigmatizing effects of their actions, or of how the health facilities’ policies or structures affect clients [3, 44, 45]. Lack of knowledge regarding the condition may also drive stigma [3, 38, 46]. For example, transmission misconceptions may drive stigmatizing, unnecessary precautions (e.g., double gloving, unnecessary quarantine), while disbelief in the curability of some stigmatized conditions may bias the provision of care [32, 35, 39]. Lacking knowledge about how to provide care for a specific condition, or lacking confidence in one’s ability to do so, may result in poor quality or discriminatory care [4, 20]. Institutional policies or systems for delivering care, such as verticalization (e.g., providing care at a separate clinic or “flagging” charts to distinguish them from the medical records of other patients) can also drive health facility stigma [3, 35].

The similarities are not only limited to drivers. The potential for generic survey tools to measure stigma (not specific to a particular health condition) was found in a literature review on leprosy, mental illness (MI), epilepsy, disability, and HIV [32]. Other studies have also found striking similarities in the consequences of stigma across diseases and cultures [15, 37, 47‐49]. In many cases, clients might experience more than one type of stigma simultaneously (e.g., HIV or tuberculosis-related stigma, or substance use stigma) [42, 50‐53].

While many health conditions are subjected to stigma, the following seven were selected as the focus of this correspondence article because of their high degree of commonality in stigma drivers: HIV, tuberculosis (TB), MI, substance abuse, diabetes, leprosy, and cancer [3, 32, 35, 39‐41]. Having a negative attitude, in particular the culpability for the condition, is a driver for all seven of these conditions, as is lack of awareness of stigma and its consequences; level of knowledge, myths, and misbeliefs; and institutional policies, procedures, and practices [3, 32, 35, 39‐43]. Fear of infection is common to four of the seven (HIV, TB, cancer, leprosy), while fear of the individual or their behavior is common to HIV, cancer, MI, and substance abuse [3, 20, 32, 33, 35, 39, 40].

In addition, although the specificities of the drivers, manifestations, and consequences of the stigmatization of different conditions can be varied (e.g., exactly what is feared), the mechanisms underlying the path between drivers, stigmatization, and its consequences often display universal characteristics. Theoretically, Link and Phelan [1] defined stigma as the co-occurrence of five components: labeling, stereotyping, separation, status loss, and discrimination [1]. The seven selected health conditions, which are stigmatized across a variety of contexts, display very similar mechanisms driving their stigmatization. Although the specific combined characteristics of a condition might be unique, the pathways through which these drivers feed the stigmatization of the seven selected conditions are often similar—especially in the specific context of health facilities.

The underlying shared mechanisms of the stigmatization process, common stigma drivers, the potential for generic health condition-related stigma measurement tools, the co-prevalence of stigmatized conditions (e.g., TB/substance abuse/HIV), and the similarities in the consequences of stigma, regardless of condition, all point to the potential for interventions to simultaneously reduce stigma related to more than one health condition at a time in health facilities. This would strengthen delivery of equitable, quality healthcare, while attending to the specific and important contextual or disease-conditions nuances.

This potential merits investigation, particularly in resource-constrained settings, where finding synergies for stigma reduction across conditions could create economies of scale, offering savings of cost and time. However, clearly, interventions must pay attention to specific cultural and socioeconomic contexts and recognize that stigmas are not always experienced in the same way in all settings.

An improved understanding of how health condition stigma is currently addressed in health facilities is needed to identify gaps and areas for investment in stigma reduction, as well as to explore the possibility of concurrently addressing more than one health condition stigma with a joint intervention. Thus, this correspondence article takes an explicitly programmatic focus and aims to examine “how” health facility-based stigma reduction interventions are implemented across health condition stigmas.

Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [54], we searched PubMed, Psychinfo and Web of Science databases in February 2018. Gray literature was obtained from the United States Agency for International Development’s (USAID) Development Experience Clearinghouse. Additionally, literature was identified through expert consultation and an ancestry citation search.

The inclusion criteria were a clear description of (a) the implementation of an intervention that aimed to reduce one of the seven health condition stigmas in healthcare settings, either by targeting the potential perpetrators of stigma (healthcare workers or healthcare facility policies) or by empowering clients to overcome stigma and discrimination and (b) the evaluation (qualitative, quantitative, process, or mixed methods) of said intervention. We strove to capture all intervention approaches and implementation methods, regardless of the target population (health workers or clients). The search was restricted to articles published in the past 5 years in English. Reviews were excluded, as were articles that only described intervention development.

Article citations and abstracts were organized, uploaded, and reviewed using EndNote. MS and KG screen abstracts to determine whether they included relevant information. The full text was obtained if at least one reviewer deemed the abstract to be relevant. MS and KG reviewed the full-text articles, and these were included if both reviewers agreed. Discrepancies were discussed with LN until a consensus was reached. Finally, MS and KG conducted ancestry searches of the citations of included articles. Data were abstracted using a standardized abstraction form adapted from a systematic review of interventions to reduce HIV-related stigma by Stangl et al. [55]. Specifically, we aimed to examine the health condition stigma addressed; the intervention populations, delivery, approaches, and methods; stigma drivers targeted; and the evaluation methods and quality.

Articles were categorized by disease-specific stigma addressed, approaches employed, intervention delivery, and stigma drivers addressed (Table 1). “Approaches” were considered as overarching strategies towards stigma reduction, and “methods” as the specific activities that reduce stigma.

MS and KG assessed the quality of quantitative data using the 27-item Downs and Black checklist [56]. Articles scoring 14 or above were considered high-quality studies [55]. The 18-item framework for evaluating qualitative evidence devised by Spencer et al. was used to assess the quality of qualitative data [57]. Studies scoring of 10 or above were considered high-quality studies [55].

A total of 728 peer-reviewed abstracts were assessed, of which 68 articles underwent full-text review and 37 met the inclusion criteria. All nine peer-reviewed records identified through a citation ancestry search were included. Forty-three gray literature records were reviewed, of which 24 underwent full-text review but none met the inclusion criteria. However, a project report identified through the ancestry search was included [58]. Forty-seven manuscripts detailing 42 distinct interventions were included (Fig. 1).

All the included interventions focused on stigma related to HIV, MI, or substance abuse. No articles meeting the inclusion criteria were found for TB, diabetes, cancer, or leprosy. Interventions that addressed more than one medical condition were only found for MI or substance abuse. Twenty of the identified interventions targeted healthcare providers, 24 targeted healthcare students, four included clients in the intervention population, and only one included all levels (medical and non-medical) of healthcare workers.

Most quantitative studies (38) scored at least 14 out of 27 points on the Black and Downs checklist and were thus categorized as high-quality studies for the purposes of this review. The scores ranged between 7 and 24, with an average score of 16.5. Over half of the interventions scored between 14 and 18 (n = 26). The two qualitative studies were categorized as being high quality (see Table 1 for individual study scores).

Interventions were implemented across the globe, with at least one intervention implemented in every World Health Organization region. The largest number (n = 16) were implemented in the Americas, eight in the USA (including one in Puerto Rico), and eight in Canada. Only one intervention was implemented in the Eastern Mediterranean. Most interventions were implemented in high-income countries (n = 27), and, of those, nearly all (n = 25) focused on MI, substance abuse, or both (Table 2). Interventions were evaluated using qualitative, quantitative, and mixed-methods (Table 1).

Several key strategies to reduce stigma in healthcare settings emerged from the reviewed interventions.

Nearly every intervention took multiple approaches to reduce stigma (n = 24), except for two purely structural integration interventions [59, 60]. The most frequently used approach was contact with the stigmatized group (n = 30), but this was closely followed by provision of information (n = 29) and participatory learning (n = 28). Limited discernable patterns emerged across geographical regions, between lower middle-income countries and higher income countries, or in how interventions combined approaches. However, contact approaches were used in combination with most of the participatory learning interventions (21 of 28) and skills-building approaches (16 of 22) (Table 3).

Different ways of implementing the various approaches described above were used. Examples of the methods that can be used by each approach can be found in Table 4. Most interventions drew on multiple approaches and, consequently, also used multiple methods to deliver those approaches. Of the non-structural interventions, they were delivered in person, using video or streaming technology, or consisted of clinical placements, rotations, or clerkships for students. Such interventions were led or delivered by professionals (e.g., professors, expert medical providers, external facilitators) or clients (i.e., members of the stigmatized group). One was led by health facility staff members who had been trained as opinion leaders to champion stigma reduction [60‐63]. Information provision approaches were delivered through didactic lectures, medical training courses, discussion, or printed educational materials. Contact approaches involved exposing the health facility staff participants to individuals living with the stigmatized condition, either in person or through videos, in non-clinical interactions. The mechanisms of these controlled exposures were through performances, discussions, participatory activities, or facilitated clinical placements. Participatory learning activities included discussion-based educational programs, interactive group work, role-playing, games, and assignments. Skills-building approaches were often operationalized through role-playing or through guided or controlled clinical practice, both with and without members of the stigmatized group.

We were unable to identify any discernable patterns of how methods or approaches were combined. Often, more passive activities, such as attending lectures or watching performances, were accompanied by open discussion or participatory activities. Of the four interventions that used structural approaches, three employed task-shifting—the redistribution of healthcare responsibilities to other sectors—and service integration. In two of these cases, HIV care was integrated into primary care, allowing HIV clients to integrate into the general patient pool and reduce their risk of status disclosure [59, 60]. Another structural intervention focused on reducing fear of HIV transmission as a means to reduce HIV-related stigma. This intervention trained facility-based stigma reduction popular opinion leaders on universal precaution procedures and provided infection protection supplies, such as gloves, to the whole facility [60‐64]. Of the MI and substance abuse interventions that used clinical placements or role-playing to provide clinical care, six focused on recovery-oriented care. Recovery-oriented care is loosely characterized by a more optimistic view of recovery, empowerment of the patient, and aligning the providers’ goals with the clients’ recovery goals [65‐70].

Four studies compared the effectiveness of different methods or approaches. Clarke et al. compared “dialectical behavioral therapy,” which aims to reduce prejudice and discrimination towards patients with personality disorders by providing staff with knowledge and skills to improve the effectiveness of their clinical practice, to “acceptance and commitment training,” which aims to provide self-management skills to reduce the impact of negative evaluations and strengthen value-driven behavior. For both types of training, staff attitudes improved and social distancing reduced, but they did not significantly differ [71]. Fernandez et al. compared the efficacy of an in-person, face-to-face contact plus educational lecture, to a video-based contact plus educational lecture. No significant differences were found between the two methods in terms of mental disorder stigma reduction [66]. Mak et al. compared the efficacy of an educational lecture plus a 90-min in-person sharing session led by people living with HIV to an educational lecture plus in-person interactive games led by research assistants (who were not living with HIV) and found no significant differences in HIV stigma reduction [72]. Winkler et al. compared an informational leaflet, a short video intervention, and a seminar involving direct contact with a mental health client. They found that attitudes and behavioral intent towards clients living with a mental disorder improved significantly for the video and seminar groups, whereas limited changes were seen for the flyer group. However, there were no significant differences between the groups [73].

Few articles explicitly identified the drivers targeted by their interventions. Li et al. targeted fear and fear-driven stigmatizing behaviors [60‐63]. Batey et al. targeted attitudes, stigma knowledge, and HIV knowledge [74]. Geibel et al. targeted health facility policies and work environment and attitudes towards sexually active young people and people living with HIV [75]. Shah et al. targeted fear of, and misconceptions about, HIV transmission and attitudes towards populations vulnerable to HIV infection [76].

Interventions targeted attitudes, knowledge of stigma, knowledge of the condition, fear, ability to clinically manage the condition, client coping mechanisms, or institutional policies (Table 1). While some interventions explicitly stated the stigma driver targeted by their intervention, others did not; in cases where the stigma drivers were not explicitly described, we inferred the drivers targeted from the overall description of the intervention. Nearly 30 interventions targeted more than one driver. The most commonly targeted driver was knowledge about the condition. No regional trends or patterns were identified.

Of the 40 unique quantitative studies, 27 reduced stigma and 13 had mixed results (Table 1). However, the included interventions were evaluated using different measures, making cross-intervention comparisons difficult. Of note, certain interventions were evaluated using a wide array of stigma measures, while others were evaluated using just a few survey questions. Some evaluations had multiple follow-up surveys, while others only used one post-intervention time-point. Others pooled their measures of stigma into an overall index or score, while others examined differences between individual items. Interventions using more stigma measures were more likely to obtain mixed results than those using just a few measures.

Several gaps emerged from the literature search. Of the 42 unique studies, most (33) focused on MI or substance abuse and nine focused on HIV. Of note was the absence of recent stigma reduction interventions in health facilities for TB, diabetes, leprosy, or cancer. This may be because the presence of health facility stigma around diabetes and cancer has only relatively recently been recognized. For leprosy, it has very low and geographically confined prevalence. Possibly, interventions are being carried out, but are not being evaluated, or results have not yet been published or were published more than 5 years ago. The dearth of evaluations of stigma reduction interventions for TB was particularly notable; the lack of interventions addressing TB stigma has been noted by two other recent reviews of TB-related stigma [77, 78].

Other gaps identified included either no or few interventions that (1) targeted all levels of clinical or non-clinical health facility staff, concentrated on multiple ecological levels, or worked to structurally change physical or policy aspects of the facility environment; (2) engaged health facility staff and clients in a collaborative effort to design and implement stigma reduction interventions; (3) leveraged technology for interactive learning beyond videos for testimonials; and (4) recognized and addressed stigma experienced by health workers.

There is growing recognition that, to deliver a sustainable and scaled response to health facility stigma, it is important to address stigma at multiple ecological levels within a health facility [3, 64, 79]. While this search of the literature identified only one intervention targeting all levels of staff in a facility [64], current efforts led by some authors of this manuscript in Thailand (the 3X4 approach) [80], Ghana, and Tanzania (the Health Policy Project total facility approach) [81] are developing and testing a package of interventions that work at both the individual (health facility staff) and structural (health facility policy and environment) levels within a facility. At the individual level, these interventions focus on participatory training of health facility staff of all cadres (clinical and non-clinical). Any health facility employee who has client contact can stigmatize; therefore, working with all cadres of health workers is important. At the structural level, the 3X4 and the Health Policy Project total facility approaches are focused on developing and enforcing anti-discrimination policies, infection control by providing supplies and enforcing standard precaution infection control practices, as well as client complaint and compliment mechanisms. Further investigation of the potential for structural interventions to reduce stigma is needed [82], particularly around how the physical layout or space within a facility can contribute to, or mitigate, the experience and anticipation of stigma in facilities [83]. Based on the experiences of staff and clients, simple physical changes can lower the experience and risk of stigma, as well as unwanted disclosure [84, 85]. For example, a pharmacist participating in the stigma reduction training in Ghana became aware that their pharmacy inadvertently stigmatized clients living with HIV (and disclosed their HIV status) by having two separate windows for medicine pick-up: one for clients living with HIV, and one for everyone else. Following the intervention, all clients now go to the same window [86, 87].

Keeping those who fear, or are burdened, by stigmatization at the center of any response to stigma has been identified as a best practice [74, 84, 85, 88]. This includes working to empower people or groups experiencing stigma, for example, by building skills and efficacy to address internalized stigma and cope with and challenge stigma, and building partnerships with gatekeepers and opinion leaders for change. From the literature identified, the most common way of involving clients experiencing stigma in the intervention was as trainers or speakers [58, 64, 67‐70, 72, 73, 76, 89‐93]. The literature search only identified one intervention that went beyond this level of engagement to focus on an “empowerment” aspect [74]. This ongoing work in Alabama, USA, brings together health workers and clients in a workshop setting outside of the facility, to share information, increase contact, and use empowerment strategies to challenge HIV-related and intersecting stigmas. The latter is done by implementing a stigma reduction project that was developed by clients and health workers. Similarly, an ongoing intervention to prevent stigma towards people with MI or substance abuse in Lima, Peru, and Toronto, Canada, brings together primary health providers and clients to reduce stigma through five steps, one of which involves providers and clients working together in creative workshops to produce art that is presented to others [94].

In recent years, healthcare systems have witnessed rapid advances in technology, including, but not limited to, the use of electronic medical records and use of the internet, tablets, and phones to provide care, collect data, and support clinical information and ongoing education. These advances, particularly the use of self-learning via tablets, the Internet, and phones offer potentially efficient methods to deliver stigma reduction to busy health facility staff [73, 95]. Technology can also offer clients a way to mitigate or avoid health facility stigma [96, 97]. An ongoing study in India has developed, and is testing, a stigma reduction intervention that targets nursing students and health facility ward staff through two self-learning sessions on tablets, and one in-person 1.5-h group session, co-led by a person living with HIV [98]. This intervention targets several key individual-level drivers of stigma, including awareness, fear, and attitudes. Another co-author is leading the ongoing Client Centered Care Coordination (C4) intervention, which uses mobile technology to empower and help clients mitigate and avoid stigma in New York state (USA), Toronto (Canada), and multiple sites in Ghana [99]. This intervention uses different phone apps to connect clients living with HIV from key population communities, to peer support and to nurses and other health personnel, and to report and receive feedback on health behaviors and illness symptoms. Using mobile apps act as an access point to health services and reduce the opportunities for being exposed to stigma in the physical space of the health facility, as well as potential unwanted disclosure of HIV status.

Lastly, we found no interventions with a specific focus on health workers living with a stigmatized disease, and addressing any stigma they may experience from co-workers or through the facility structures. Research has shown that stigma affects healthcare workers, either because of their own health status or as a result of working with stigmatized individuals [100, 101]. The HaTSaH study, an ongoing study in Free State province, South Africa, is addressing this gap through a combination intervention approach that focuses on reducing HIV and TB stigma among health workers towards colleague health workers living with HIV and TB through clinical, structural, and sociobehavioral factors [102].

Across these sets of ongoing efforts, which address different health condition stigmas, several factors are being recognized as key to the interventions. Involvement of clients living with the stigmatized condition or behavior is critical, whether this is by creating safe spaces for contact (e.g., panel discussions), as trainers, or as participants in joint provider–client workshops. It is critical to build facility management buy-in and ownership, while also creating and empowering facility-based “champion” teams of health facility workers and clients who develop and lead tailored stigma reduction efforts in their facilities. Additionally, it is important to pay attention to physical space and how it can lead to stigma and or unwanted disclosure of status.

Future investment in research and health facility stigma reduction interventions should: