Association between clinical encounter frequency and HIV-related stigma among newly-diagnosed people living with HIV in Rwanda

Abstract

HIV-related stigma in healthcare settings remains a key barrier to engaging people living with HIV (PLHIV) in care. This study investigated the association between clinical encounter frequency and HIV-related anticipated, enacted, and internalized stigma among newly-diagnosed PLHIV in Rwanda. From October 2020-May 2022, we collected data from adult PLHIV on antiretroviral therapy (ART) in Kigali, Rwanda who were participating in a randomized, controlled trial testing early entry into differentiated care at six months after ART initiation. We measured anticipated HIV stigma with five-point Likert HIV Stigma Framework measures, enacted stigma with the four-point Likert HIV/AIDS Stigma Instrument, and internalized stigma with the four-point Likert HIV/AIDS Stigma Instrument. We used multivariable linear regression to test the associations between clinical encounter frequency (average inter-visit interval ≥50 days vs < 50 days) and change in mean anticipated, enacted and internalized HIV stigma over the first 12 months in care. Among 93 individuals enrolled, 76 had complete data on encounter frequency and stigma measurements and were included in the present analysis. Mean internalized stigma scores of all participants decreased over the first 12 months in care. Anticipated and enacted stigma scores were low and did not change significantly over time. There was no association between encounter frequency and change in internalized stigma. In this pilot study of newly-diagnosed Rwandan PLHIV with relatively low levels of HIV- related stigma, clinical encounter frequency was not associated with change in stigma. Additional research in diverse settings and with larger samples is necessary to further explore this relationship.

Introduction

HIV-related stigma - negative attitudes and beliefs about people living with HIV (PLHIV) - remains a key barrier to engaging in care [1–3]. HIV-related stigma is associated with poor mental health [4–5], hesitancy to undergo HIV testing [6] or disclose HIV serostatus [7], poor-retention in HIV care [3], and decreased adherence to antiretroviral therapy (ART) [8]. Prior studies among PLHIV in Rwanda, other Sub-Saharan African countries and globally have demonstrated that engaging with HIV care can be a stigmatizing experience for PLHIV, where fear of discrimination and unintentional HIV status disclosure in clinical settings is compounded by frequent clinical encounters and long wait times in health settings [2, 9–13]. To improve care engagement, overall health and quality of life for PLHIV, the 2025 UNAIDS targets call for less than 10% of PLHIV experiencing stigma and discrimination [14].

In order to tailor HIV treatment and reduce the burden of care on both patients and health systems, differentiated service delivery (DSD) was recommended by the WHO in 2016 as a strategy to meet the diverse preferences and needs of PLHIV [15]. DSD is an approach towards HIV care that is centered on client needs, adapting care delivery across the cascade to reflect varying needs and preferences of PLHIV. Evidence has shown that health facility-based DSD models allowing fewer clinical encounters are feasible, acceptable and effective with respect to retention in care and viral suppression, with the added benefits of improving client satisfaction and decreasing financial expenditures [16–19]. Because engaging in care can be stigmatizing, DSD has the potential to reduce HIV-related stigma by decreasing exposure to the clinical setting. To our knowledge, however, no studies have longitudinally investigated the impact of DSD models, particularly those offering infrequent clinical encounters on HIV-related stigma.

Therefore, we investigated whether clinical encounter frequency was associated with change in anticipated, enacted and internalized HIV-related stigma among newly-diagnosed PLHIV living in Kigali, Rwanda. We hypothesized that PLHIV who had less exposure to the healthcare setting (infrequent clinical encounters) would have lower levels of HIV-related stigma.

Methods

Study Design

This analysis uses data from a longitudinal, unblinded pilot randomized control trial (RCT) (NCT04567693) examining the feasibility, acceptability and preliminary effectiveness of reducing time to entry to DSD from 12 to 6 months after ART initiation as well as reducing from two to one the number of suppressed viral loads required to enter DSD among newly-diagnosed PLHIV in Kigali, Rwanda [20]. The study was conducted from October 2020 – May 2022 and was approved by the Rwanda National Ethics Committee (0918/RNEC/2020) and by the institutional review board of Albert Einstein College of Medicine (2020–11551). All participants provided written informed consent prior to enrollment in the study.

Setting

Rwanda is a landlocked nation in East Africa with a population of 13 million, HIV incidence of 0.08% and HIV prevalence of 3.0% (2018–2019) [21–22]. This study was carried out in three public health facilities located in Kigali, Rwanda’s capital, where both HIV incidence (0.11%) and prevalence (4.3%) are higher than national averages (2020–2021) [22]. The three study facilities are located in three distinct neighborhoods in Kigali, with each providing primary care (health promotion, preventive and treatment services in medicine, surgery, obstetrics and pediatrics) for acute and chronic health conditions to adult and pediatric populations in their catchment area. Primary HIV care – including ART delivery, prevention of mother-to-child transmission, male circumcision, social services and peer support – are delivered to nearly 6,000 PLHIV across the three health facilities, including approximately 300 newly-diagnosed and newly-enrolled PLHIV annually [21–22]. At the time of this study, national guidelines required that newly-diagnosed PLHIV have at least monthly clinical encounters with appointments every 3 months and medication refills every month for at least 12 months in care; patients virally suppressed at 12 months were eligible for advancement to a DSD model allowing quarterly clinical encounters with biannual appointments and ART refills every 3 months [23].

Participants

Participants were newly-diagnosed (within 6 months), adult (≥15 years) PLHIV on ART who enrolled in HIV care at a participating study health center within the 30 days prior to study enrollment. In accordance with national guidelines that limited reduced-frequency DSD models to certain populations, we excluded participants who at the time of study enrollment were pregnant or lactating, diagnosed with active tuberculosis, or diagnosed with a concurrent severe mental health or substance use disorder. We also excluded participants who were planning to move away from Kigali during the 12-month study duration or who were unable to provide informed consent.

Data Sources and Collection

The baseline visit occurred within 30 days of a participant’s enrollment in HIV care, with additional research visits at 6 and 12 months after enrolling in HIV care. During research visits, participants completed structured interviews in Kinyarwanda by trained research staff who read questions to each participant and then input responses into a REDCap study database. Participants were compensated 8000 RWF (~$8 USD) for every research visit. We extracted data on HIV diagnosis date, ART initiation date, and clinical encounter dates from participants’ medical records.

Exposure – Frequency of Clinical Encounters

We compared participants with infrequent clinical encounters because they advanced to early DSD with those who experienced frequent clinical encounters because they remained in usual care. At enrollment, participants were randomized to one of three arms in a 1:1:1 ratio: (1) entry into DSD at six months after ART initiation, after one suppressed viral load; (2) entry into DSD at six months after ART initiation, after two suppressed viral loads; (3) usual care. For the first six months of care, all participants were scheduled for quarterly appointments and monthly ART refills according to guideline-based care. Participants randomized to early DSD and who were virally suppressed were eligible to advance to DSD at six months after ART initiation, although health center clinicians were able to override study assignment.

Because a substantial number of participants randomized to early DSD did not ultimately advance to this model of care, in order to assess the association between encounter frequency and stigma we categorized participants as having infrequent or frequent clinical encounters as follows. We first determined an eligibility for DSD date for each participant, defined as the first clinical encounter occurring ≥150 days after ART initiation (to account for participants whose 6-month clinical encounter occurred earlier than scheduled). We then calculated the average interval between each clinical encounter from the eligibility for DSD date until the end of the study for each participant. Participants whose average interval between clinical encounters was ≥50 days were categorized as having infrequent encounters, while those whose average interval between encounters was <50 days were categorized as having frequent encounters.

Outcomes

The primary outcome measures were anticipated, enacted, and internalized HIV stigmas, which were defined based on Earnshaw and Chaudoir’s HIV Stigma Framework [24]. These three forms of HIV stigma are distinct concepts that capture the complex mechanisms that shape how PLHIV experience stigma and engage with HIV care [26–28]. All stigmas were measured as ordinal variables and scores were determined by averaging participant’s numerical responses to each discrete set of questions associated with each type of HIV stigma. Anticipated HIV stigma (i.e. expectations of future discrimination and/or prejudice due to one’s HIV status) was assessed using measures from the HIV Stigma Framework [29] by asking participants a series of six questions about anticipated negative treatment from family members and healthcare providers. Responses were measured on a scale of 1 (very unlikely) to 5 (very likely). Enacted stigma (e.g. previous or current experiences of discrimination and/or prejudice from others due to one’s HIV status) [25] was measured using a series of 28 questions from the HIV/AIDS Stigma Instrument [29] asking participants to quantify experiences over the past three months that involved prejudice, discrimination, and/or negative treatment towards them as the result of their HIV status. These questions covered enacted stigma experienced both in social and community settings as well as in healthcare settings. Responses were rated on a scale of 1 (never) to 4 (most of the time). Internalized HIV stigma (e.g. attributing perceived community held beliefs to oneself [3], resulting in the development of negative internal representations that may lead to demoralization, diminished self-efficacy and emotional distress) [5], [6] was measured using the HIV/AIDS Stigma Instrument [29]. Internalization was measured through a series of five questions that asked participants how many times in the past three months they had negative thoughts or feelings related to their HIV status. Responses were also rated on a scale of 1 (never) to 4 (most of the time).

Other Measures

We collected baseline demographic characteristics including participant health center, age, gender (man, woman, other), sexual orientation (straight, gay/lesbian/queer), level of education completed (none, primary, secondary, vocational, university), marital status (single, married, other), who the patient lived with (alone, with spouse, other family members, or roommates, or without housing), employment status (unemployed, self-employed, part-time employed, full-time employed), household income (RWF) and health insurance status. Clinical measures included time from HIV diagnosis to ART initiation, time from HIV care enrollment to ART initiation, and self-reported overall health on the given day of the survey using the EuroQOL VAS tool [30], where participants rated their overall health from 0 to 100, where 100 indicated the best health that a respondent could imagine.

Analysis

We first compared baseline characteristics of participants with infrequent and frequent encounters using chi-squared tests or Fisher’s exact test for categorical variables and t-tests for continuous variables. For each participant, we calculated anticipated, enacted, and internalized HIV-related stigma scores at baseline, 6 months, and 12 months. We compared mean within group stigma from baseline to 12 months and 6 to 12 months using paired t-tests to assess significance of change in each type of stigma over time.

We used bivariate and multivariable linear regression to test the association between encounter frequency and mean change in internalized stigma between 6 and 12 months. We constructed a multivariable linear regression model to assess any impact on the association between encounter frequency and decrease in internalized stigma when adjusting for variables (age group, sex, education level, employment status, baseline quality of life score) that were known have an impact on HIV-related stigma based on prior evidence [31–33]. Because we did not observe any significant change in mean anticipated and enacted stigma over time within each group or among the overall sample, we did not perform further analyses for these outcomes.

Results

Of 93 enrolled participants, three were withdrawn from the study because they did not meet inclusion criteria, two died during the study, four were lost to follow-up, and eight did not have complete encounter data available. This analysis includes N=76 participants, the majority of whom were women (67.1%), with a mean age of 31 years. Half of all participants (50.0%) completed primary school, 14.5% secondary school, and 6.6% completed university. The median monthly household income of participants was 30,000 RWF ($24.54 USD), 71.1% reported some type of employment and 76.3% had health insurance coverage. The average baseline quality of life score among participants was 43.3 on a scale from 0–100 (Table 1). Among included participants, 36 (47.4%) had infrequent clinical encounters and 40 (52.6%) had frequent encounters.

Table 1.

Baseline Demographic and Clinical Characteristics

	All Participants N=76 n (%)	Infrequent Encountersa (N=36) n (%)	Frequent Encountersb (N=40) n (%)	Test Statistic	p
Health Center				1.00c	0.61
Gikondo	31 (40.8)	15 (41.7)	16 (40)
Kicukiro	14 (18.4)	5 (13.9)	9 (22.5)
Remera	31 (40.8)	16 (44.4)	15 (37.5)
Age Group				0.08c	0.78
15–23	20 (26.3)	10 (27.8)	10 (25)
>23	56 (73.7)	26 (72.2)	30 (75)
Age (years) Mean (SD)	31.2 (10.2)	31.1 (11.0)	31.1 (10.0)	0.00d	0.99
Gender				0.81c	0.37
Male	25 (32.9)	10 (27.8)	15 (37.5)
Female	51 (67.1)	26 (72.2)	25 (62.5)
Sexual Orientation				0.40e	1.00
Straight	72 (96)	34 (97.1)	38 (95)
Gay/Lesbian/Queer	3 (4)	1 (2.9)	2 (5)
Education Completed				0.001e	0.41
None	16 (21.1)	6 (16.7)	10 (25)
Primary School	38 (50)	17 (47.2)	21 (52.5)
Technical School	6 (7.9)	5 (13.9)	1 (2.5)
Secondary School	11 (14.5)	5 (13.9)	6 (15)
College or University	5 (6.6)	3 (8.3)	2 (5)
Marital Status				0.04e	0.48
Single	63 (82.9)	32 (88.9)	31 (77.5)
Married	8 (10.5)	2 (5.6)	6 (15)
Other (divorced, widowed, separated)	5 (6.6)	2 (5.6)	3 (7.5)
Who do you live with?				0.02e	1.00
Alone	17 (22.4)	8 (22.2)	9 (22.5)
With husband/wife or romantic partner	21 (27.6)	10 (27.8)	11 (27.5)
With other family members or roommates	34 (44.7)	16 (44.4)	18 (45)
Other or No place to stay	4 (5.3)	2 (5.6)	2 (5)
Employment Status				0.004e	0.39
Unemployed	22 (28.9)	12 (33.3)	10 (25)
Self-employed	6 (7.9)	1 (2.8)	5 (12.5)
Part-time employed	36 (47.4)	16 (44.4)	20 (50)
Full-time employed	12 (15.8)	7 (19.4)	5 (12.5)
Median Household Income (RWF)	30,000	35,000	30,000	0.63f	0.53
Health Insurance				0.63c	0.43
Yes	58 (76.3)	26 (72.2)	32 (80)
No	18 (23.7)	10 (27.8)	8 (20)
Baseline Quality of Life VAS	43.3 (15.7)	46.4 (14.6)	39.8 (15.5)	3.64d	0.06
Median time from HIV care enrollment to ART Initiation	0	0	0	0.37f	0.71
Time from HIV Diagnosis to ART Initiation				0.05e	0.89
Same Day	46 (60.5)	21 (58.3)	25 (62.5)
1–7 Days	23 (30.3)	12 (33.3)	11 (27.5)
>7 days	7 (9.2)	3 (8.3)	4 (10)

^aInfrequent Encounters: Inter-encounter interval >50 days

^bFrequent Encounters: Inter-encounter interval ≤50 days

^cChi-squared test

^dF-value

^eFisher’s Exact Test

^fMedian Two-Sample Test

At baseline the mean anticipated stigma score was 2.41, corresponding with participants’ belief that it would be unlikely for them to be treated negatively by either family members or healthcare providers due to their HIV status. At 6 months the mean anticipated stigma score was 2.49, and at 12 months it was 2.41, with no significant change across time points (6 to 12 month difference −0.09; 95% CI −0.23 – 0.06, t=−0.58, p=0.24) (Table 2, Figure 1). At baseline the mean enacted stigma score was 1.03, corresponding to having never experienced HIV-related stigma in the 3 months prior to the survey. At 6 months mean enacted stigma score was 1.01, and at 12 months 1.03, with no significant change across all time points (6 to 12 month difference 0.02; 95% CI −0.01 – 0.05, t=1.50, p=0.14) (Table 2, Figure 1). There was no significant change over time in anticipated or enacted stigma among either participants with infrequent encounters or those with frequent encounters.

Table 2.

Change in HIV Anticipated, Enacted, and Internalized Stigma Scores

	Baseline Score Mean (SD)	6 Month Score Mean (SD)	12 Month Score Mean (SD)	Baseline to 12mo Mean Difference (SD)	95% CI of Difference (Lower - Upper)	t Statistic	p	6mo to 12mo Mean Difference (SD)	95% CI of Difference (Lower - Upper)	t Statistic	p
ANTICIPATED STIGMA
All Participants (N=76)	2.41 (0.76)	2.49 (0.62)	2.41 (0.62)	−0.002 (0.82)	−0.18 – 0.18	−0.33	0.97	−0.09 (0.66)	−0.23 – 0.06	−0.58	0.24
Infrequent Encounters a (N=36)	2.45 (0.73)	2.44 (0.53)	2.44 (0.66)	−0.01 (0.76)	−0.27 – 0.24	−0.07	0.94	−0.005 (0.71)	−0.24 – 0.23	−0.04	0.97
Frequent Encounters b (N=40)	2.40 (0.81)	2.43 (0.66)	2.35 (0.62)	−0.05 (0.88)	−0.33 – 0.23	−0.36	0.72	−0.08 (0.63)	−0.28 – 0.12	−0.80	0.43
ENACTED STIGMA
All Participants (N=76)	1.03 (0.13)	1.01 (0.05)	1.03 (0.15)	0.003 (0.21)	−0.04 – 0.05	0.13	0.91	0.02 (0.13)	−0.01 – 0.05	1.50	0.14
Infrequent Encounters a (N=36)	1.05 (0.2)	1.00 (0)	1.03 (0.18)	−0.02 (0.28)	−0.12 – 0.07	−0.52	0.61	0.03 (0.18)	−0.03 – 0.09	1.00	0.32
Frequent Encounters b (N=40)	1.00 (0.01)	1.01 (0.08)	1.03 (0.14)	0.03 (0.14)	−0.02 – 0.07	1.23	0.23	0.02 (0.07)	−0.006 – 0.04	1.51	0.14
INTERNALIZED STIGMA
All Participants (N=76)	1.89 (0.83)	1.35 (0.68)	1.14 (0.45)	−0.75 (0.89)	−0.94 – −0.55	−7.18	<0.001	−0.20 (0.69)	−0.35 – −0.05	−2.55	0.01
Infrequent Encounters a (N=36)	1.94 (0.81)	1.37 (0.82)	1.22 (0.56)	−0.07 (0.97)	−1.05 - −0.39	−4.47	<0.001	−0.15 (0.86)	−0.44 – 0.14	−1.05	0.30
Frequent Encounters b (N=40)	1.80 (0.86)	1.29 (0.52)	1.05 (0.20)	−0.75 (0.83)	−1.01 – −0.48	−5.70	<0.001	−0.24 (0.46)	−0.39 – −0.09	−3.33	0.002

^aInfrequent Encounters: Inter-encounter interval >50 days

^bFrequent Encounters: Inter-encounter interval ≤50 days

Figure 1. Change in HIV-related anticipated, enacted, internalized stigma over the first 12 months in care.

Comparison of participants with infrequent clinical encounters (inter-appointment interval >50 days) versus those with frequent clinical encounters (inter-appointment interval ≤50 days).

At baseline the mean internalized stigma score was 1.89, corresponding to experiencing internalized stigma once or twice in the past 3 months. At 6 months the mean internalized stigma score was 1.35 and at 12 months was 1.14, which corresponds with almost never experiencing internalized stigma in the past 3 months. We observed a significant decrease in the mean internalized stigma score of all participants from baseline to 12 months of −0.75 (95% CI −0.94 – −0.55, t=−7.18, p <0.001) and from 6 months to 12 months of −0.20 (95% CI −0.35 – −0.05, t=−2.55, p=0.01). Between 6 and 12 months, internalized stigma did not significantly decrease among participants with infrequent encounters (−0.15, 95% CI −0.44 – 0.14, t=−1.05, p=0.30), but did significantly decrease among participants with frequent encounters (−0.24, 95% CI −0.39 – 0.09, t=−3.33, p=0.002). However, the association between change in internalized stigma and clinical encounter frequency was not significant in the bivariate linear regression (beta=0.09, 95% CI −0.22 – 0.40, p=0.56) or in a multivariable regression controlling for age, sex, education level, employment status, and baseline quality of life score (Table 3).

Table 3.

Multivariable Model for Difference Between Groups of Change in Internalized Stigma from 6–12 Months

	Baseline Score Mean (SD)	6 Month Score Mean (SD)	12 Month Score Mean (SD)	Linear Regression Coefficient	95% CI of Difference (Lower - Upper)	t Statistic	p
Encounter Frequency
Infrequent Encountersa	1.94 (0.81)	1.37 (0.82)	1.22 (0.56)	0.13	−0.18 – 0.44	−0.82	0.42
Frequent Encountersb	1.80 (0.86)	1.29 (0.52)	1.05 (0.20)	Ref.	-
Age Group
Age 15–23	2.17 (0.82)	1.41 (0.80)	1.10 (0.44)	Ref.	-
Age >23	1.79 (0.82)	1.32 (0.64)	1.16 (0.45)	−0.08	−0.44 – 0.29	−0.41	0.68
Sex
Female	2.02 (0.84)	1.45 (0.76)	1.16 (0.47)	Ref.	-
Male	1.63 (0.76)	1.14 (0.45)	1.11 (0.40)	0.27	−0.09 – 0.63	1.52	0.30
Education Level
No schooling or Completed Primary School	1.92 (0.88)	1.33 (0.67)	1.15 (0.46)	Ref.	-
Completed technical, secondary school, or college/university	1.81 (0.68)	1.39 (0.72)	1.11 (0.43)	−0.20	−0.56 – 0.15	−1.14	0.26
Employment Status
Unemployed/self-employed	1.70 (0.80)	1.48 (0.83)	1.11 (0.44)	Ref.	-
Full/part-time employed	2.01 (0.84)	1.25 (0.56)	1.16 (0.46)	0.26	−0.07 – 0.59	1.55	0.12
Baseline Quality of Life (VAS Score)				0.01	−0.004 – 0.02	1.35	0.18

^aInfrequent Encounters: Inter-encounter interval >50 days

^bFrequent Encounters: Inter-encounter interval ≤50 days

Discussion

This is among the first studies of PLHIV to longitudinally investigate the relationships between clinical encounter frequency and change in HIV-related stigmas. We observed a decrease in internalized stigma over the first 12 months in care; however, both anticipated and enacted stigma were uniformly low and remained so over the first 12 months in care. Additionally, we observed no significant associations between change in any type of HIV-related stigma with encounter frequency, in contrast to our hypothesis that PLHIV with infrequent clinical encounters would have lower levels of HIV-related stigma.

While we found that HIV-related internalized stigma decreased over the first 12 months in care for all participants, this decrease was not significantly associated with the frequency of encounters. This was unexpected given prior studies that have reported stigma and fear of discrimination among PLHIV when exposed to the healthcare setting [2, 10–13, 39, 44–47]. As nearly all participants initiated ART within 1 week of HIV diagnosis, it is possible that early engagement in care, which includes supportive counseling, may have a protective effect against internalized stigma, even in the setting of frequent encounters [36–38]. Other potential explanations for the observed decrease in internalized stigma, regardless of encounter frequency, include improved physical health due to ART initiation and gradual acceptance of one’s HIV diagnosis over time [38].

We also found low and unchanging levels of both anticipated and enacted stigma over the first 12 months in care. On average, participants reported that they were unlikely to anticipate negative treatment or discrimination from family members or healthcare providers because of their HIV status. Almost all participants reported never experiencing enacted HIV-related stigma in the 3 months prior to the baseline, 6-month, and 12-month surveys. Low levels of anticipated and enacted stigma during baseline and even 6-month surveys may also be due to the short interval since diagnosis and relatively limited time to experience HIV stigma. Overall, these data differ from many reports describing persistent stigma in community and healthcare settings globally [12, 39–40]. It is possible that the low levels of anticipated and enacted stigma observed among participants are a result of the successful implementation of “Treat All” in Rwanda and subsequent expansion of access to HIV testing and treatment, linkage to care, and development of HIV education, awareness, prevention and anti-stigma initiatives [41–43]. These findings may also be the result of social desirability bias resulting in participant hesitancy to report anticipated and enacted stigma. We found no difference in anticipated or enacted stigma scores between participants with infrequent or frequent encounters, which further refuted our hypothesis that fewer clinical encounters would result in lower levels of HIV-related stigma.

This study had several limitations. First, this is a relatively small pilot study, with some participants who did not have complete encounter data; as such, this analysis may have been underpowered to detect differences in stigma by frequency of clinical encounters. Given the pilot nature of this study, we did not have the resources to carry out validation of the stigma scales in the specific study population, nor were we able to undertake a process of cognitive interviewing to ensure that the questions utilized to measure stigma were interpreted as they were intended. We recognize that this limited our ability to be certain that the questions used to measure each type of stigma were understood by our study participants in ways similar to other settings. The instruments that we used to assess each type of stigma were not validated in Kinyarwanda which may also contribute to misinterpretation of the questions among participants and may impact the reliability and validity of the results. Although the measures we utilized have been validated in several neighboring sub-Saharan African countries [48–52], larger-scale research efforts in this area would benefit from specific validation efforts in Rwanda. It is possible that social desirability biases may have contributed to participant reluctance to report the true extent of anticipated, enacted, and internalized stigma. Further, we conducted this study in Kigali, Rwanda, a city with a strong HIV care network and public health system; thus results from this study may be limited in their generalizability. Participants also predominantly identified as straight and the majority were over the age of 23 years. As such, we were limited in our understanding of how these findings apply to younger PLHIV and to people who identify with other sexual orientations, who may experience HIV-related stigma differently as the result of other lived experiences.

Conclusion

In conclusion, among newly diagnosed PLHIV in Kigali, Rwanda, while internalized stigma decreased in the first 12 months of care, encounter frequency was not associated with this change. We also observed low and unchanging levels of anticipated and enacted stigma among our participants and over time. It is likely that there are multiple mechanisms driving the observed decrease in internalized stigma and that the 2016 implementation of Treat All in Rwanda has mediated greater access to care and social support for PLHIV resulting in generally low levels of anticipated and enacted stigma. Attention must be paid to understanding the intersections of HIV-related stigma and other forms of public, professional, institutional, and self-stigmas. Future studies undertaken in Rwanda should ensure that HIV stigma scales utilized undergo population-specific validation. Additionally, differentiated care models must be further investigated in larger cohorts to better understand their impacts and to more readily design models that can meet the variable needs of all people living with HIV, at all points along the care continuum.

List of Abbreviations

ART

Antiretroviral therapy

DSD

Differentiated service delivery

HIV/AIDS

Human immunodeficiency virus/Acquired immunodeficiency syndrome

IRB

Institutional review board

PLHIV

People living with HIV

RWF

Rwandan Franc

WHO

World Health Organization