Understanding Oral Prep Interest Among South African Adolescents: The Role of Perceived Parental Support and PrEP Stigma

Abstract

We examined oral PrEP interest among adolescents and its association with perceived parental support and PrEP stigma. Cross-sectional data were collected during baseline procedures of the “Our Family Our Future” intervention trial in South Africa. Adolescents (14–16 years) at elevated risk for acquiring HIV and their parents or caregivers were dyadically enrolled from 2018 to 2021. There were 879 complete adolescent-parent dyads. Among adolescents, 27% had heard about PrEP, 67% reported they would want to use PrEP, and 58% thought their parent would want them to use PrEP. Among parents, 33% had heard about PrEP and 85% reported they would want their adolescent to use PrEP. Adolescents who thought their parent would want them to use PrEP were more likely to be interested in PrEP than adolescents who thought their parent would not want them to use PrEP (adjusted prevalence ratio (aPR) = 2.11, 95% CI 1.82, 2.44). Further, adolescents with higher average PrEP stigma scores above the adolescent sample median were less likely to be interested in PrEP than adolescents with lower average PrEP stigma scores (aPR = 0.81, 95% CI 0.72, 0.91). In conclusion, parents were more supportive of their adolescent taking PrEP than adolescents perceived they would be, and perceptions of low parental support and greater PrEP stigma were associated with reduced PrEP interest among adolescents. Interventions should aim to improve adolescent-parent communication around sexual health and effective HIV prevention tools.

Introduction

Adolescents and young adults aged 15–24 years account for over one-third of all new global infections of HIV, with the vast majority of infections occurring in sub-Saharan Africa (SSA) [1]. Oral antiretroviral pre-exposure prophylaxis, or PrEP, is a biomedical HIV prevention strategy that holds enormous potential to reduce HIV acquisition by greater than 90% when adherence is high [2–4]. Recently, PrEP demonstration projects in the United States and South Africa have shown that PrEP is safe and well-tolerated among adolescents under 18 years of age [5–7]. As PrEP rollout expands to include adolescents globally, a lack of parental support and PrEP-related stigma have emerged as barriers to PrEP uptake and use in qualitative research [8–13].

Parental support may be an especially important factor to consider with respect to adolescent PrEP use. Parental permission laws in certain settings may prevent minors from accessing HIV prevention services without parent or caregiver consent [14]. Further, since most adolescents live at home, parents can have a strong influence on adolescent motivations, decisions, and behaviors relating to health [15]. Research among South African adolescent girls ≤ 18 years has shown that those who disclosed their PrEP use to a parent were more likely to have high adherence after six months on PrEP compared to those who did not [16]. Further, in a study investigating PrEP interest among adolescent men who have sex with men (MSM) in the United States, perceptions of low parental support contributed to reduced interest in PrEP [8]. The parent-child interaction around PrEP interest in SSA remains largely unexplored.

Stigma has emerged as a significant social harm that can arise from PrEP use in acceptability studies [17–20] and has been identified by adolescents as a barrier to both PrEP uptake and use in qualitative research [8–12]. PrEP use generates risk for multiple stigmas, including the stigma of being related to HIV and participating in high-risk behaviors [21]. PrEP use can also evoke concerns that one is breaking social norms or putting oneself at risk for violence by engaging in sexual activity or taking PrEP [12]. For younger adolescents interested in PrEP, concerns surrounding stigma may be particularly pronounced. The association between PrEP stigma and interest among adolescents in SSA has not yet been quantified.

We surveyed South African adolescents (14–16 years) at an elevated risk of acquiring HIV and their parents or caregivers to assess adolescent interest in oral PrEP, adolescent perceptions of their parents’/caregivers’ support for their potential PrEP use, parent/caregiver support for their adolescent’s potential PrEP use, and perceptions of PrEP stigma among both dyad members. We examined associations between (1) perceived parental/caregiver support and PrEP interest among adolescents and (2) level of PrEP stigma and interest in/support for adolescent PrEP use among adolescents and parents/caregivers, respectively.

Methods

Study Design

We integrated a cross-sectional survey into the baseline procedures of the “Our Family Our Future” resilience-oriented family intervention trial (NCT03231358). The Our Family Our Future intervention integrates two empirically supported programs to prevent adolescent HIV, sexually transmitted infections, and depression [22]. The trial enrolled male and female adolescents (14–16 years) at high risk for acquiring HIV with moderate depressive symptoms and their parents or caregivers from August 2018 through August 2021. Eligible adolescent-parent or adolescent-caregiver dyads were randomized (1:1) to receive the family intervention or standard of care and followed for one year. At baseline, before trial randomization, both dyad members completed separate surveys administered via audio computer-assisted self-interview (ACASI) software. Surveys included questions evaluating oral PrEP awareness, interest in and support for adolescent oral PrEP use, and perceptions of PrEP stigma. All procedures were approved by the University of Cape Town Human Research Ethics Committee (REF 374/2017). Written informed assent and parent/legal guardian consent were obtained for adolescent participation, and written informed consent was obtained for parent/ caregiver participation.

PrEP was not offered to adolescents as part of this study. PrEP is freely available in government clinics in South Africa, and parental consent is not required for adolescents < 18 years to obtain PrEP. However, at the time of this study, PrEP rollout and training procedures were still ongoing, and PrEP was not being actively offered to adolescents in most clinics.

Study Population

Participants were recruited via systematic door-to-door sampling of households located in a peri-urban settlement of approximately 25,000 people located 40 km south of Cape Town, South Africa. The HIV prevalence among adults in this community is approximately 25%. Each housing unit within community census enumeration areas was contacted up to three times to assess if dyads were interested and met eligibility criteria. Adolescents were eligible if they (1) were 14–16 years, (2) had moderate symptoms of depression (score 6–15 inclusive) using the Center for Epidemiologic Studies Depression scale for Children (CES-DC), and (3) could identify a parent or primary caregiver with whom they lived in the same household at least four days per week. Parents/caregivers were excluded if they were < 18 years or if they reported clinically significant thresholds of depression (score ≥ 16) on the adult CES-D. The term “parents” will be used to describe both parents and caregivers throughout this manuscript.

Data Collection

During their respective baseline ACASI surveys, adolescents and parents were given a brief description of oral PrEP (“oral PrEP is a pill taken by HIV-negative people every day to protect them from getting HIV”). Then, to assess PrEP awareness, all participants were asked: “Have you heard of oral PrEP before today” (yes/no). Adolescent participants were then asked about PrEP interest (“How would you feel about using oral PrEP to prevent HIV?”) and perceived parental support (“How do you think your parent would feel about you using oral PrEP to prevent HIV?”). Each PrEP acceptability question was rated on a 5-point Likert scale ranging from “[I/My parent] would definitely want [me] to use PrEP” to “[I/My parent] would definitely NOT want [me] to use PrEP.” Parent participants were asked to rate: “How would you feel about your child using oral PrEP to prevent HIV?” on a 5-point Likert scale ranging from “I would definitely want my child to use PrEP” to “I would definitely NOT want my child to use PrEP.”

Lastly, both dyad members received a 7-item PrEP stigma measure, of which questions were created from prior PrEP acceptability research in the South African context [12, 18]. For adolescent participants, measure items were framed to assess perceptions of anticipated PrEP stigma that they might experience if they were to take PrEP. For parents, measure items were framed to assess perceptions of anticipated PrEP stigma that they themselves (i.e., the parents) might experience if their child was taking PrEP. Each question was rated on a 5-point Likert scale ranging from “strongly disagree” to “strongly agree.” The PrEP stigma measure had good internal consistency for both adolescent (Cronbach’s α = 0.85) and parent (α = 0.91) scales. Full measures for adolescents and parents can be found in the Supplemental material

Analysis

We conducted a descriptive and bivariate analysis of oral PrEP awareness and interest, as well as perceived parental support and actual parental support for adolescent oral PrEP use among matched adolescent-parent dyads. Dyads were excluded if the adolescent self-reported that they were living with HIV since questions were related to HIV prevention. For adolescents, we described the proportion of participants who (1) reported they had heard of PrEP previously, (2) would “definitely” or “probably” want to use PrEP, and (3) who thought their parent would “definitely” or “probably” want them to use PrEP. For parents, we described the proportion of participants who (1) reported they had heard of PrEP previously and (2) would “definitely” or “probably” want their child to use PrEP. Log-binomial regression was used to estimate the relative prevalence of PrEP interest among adolescents who thought their parent would “definitely” or “probably” support their potential PrEP use vs. those who thought their parent would “definitely not” or “probably not” support their potential PrEP use or were “unsure.”

Next, we described the proportion of adolescents and parents who agreed or strongly agreed with each PrEP-stigma measure item. Average PrEP stigma scores (range = 1–5) were dichotomized at the adolescent and parent sample medians for participants with complete data for PrEP stigma items. We then estimated (1) the relative prevalence of PrEP interest among adolescents with higher vs. lower PrEP stigma scores and (2) the relative prevalence of support for adolescent PrEP use among parents with higher vs. lower PrEP stigma scores. Adolescent models adjusted for prior PrEP awareness, gender, and lifetime history of vaginal or anal sex; parent models adjusted for prior PrEP awareness, gender, and self-reported HIV status.

Participants who refused to answer PrEP questions (i.e., declinations) were considered to have missing data. We used multiple imputation to assess the sensitivity of estimates derived from our primary complete-case analysis to missing values. Among adolescents, 135 (15%) refused to answer if they would want to use PrEP, 130 (15%) refused to answer if they think their parent would want them to use PrEP, and 202 (23%) refused at least one PrEP stigma measure item. Among parents, 226 (26%) refused to answer if they would want their child to use PrEP and 248 (28%) refused at least one PrEP stigma measure item. Chi-square tests were used to identify prior PrEP knowledge and gender as covariates significantly related to missingness of PrEP measures. Forty imputed datasets were created using the fully conditional specification [23] with iterative multinomial logistic imputation for missing PrEP responses.

Results

The study enrolled 953 matched adolescent-parent dyads who completed baseline interviews. Among these, 74 adolescents self-reported that they had previously tested positive for HIV (i.e., were known to be living with HIV), and these dyads were excluded from the analytic sample. Adolescents (n = 879) were 53% female and were almost all currently attending school (95%). Just less than half (45%) had engaged in vaginal or anal intercourse in their lifetime, of whom 27% reported never having used a condom. Parents (n = 879) had a median age of 36 years (interquartile range (IQR) = 27–42 years) and were 90% female. Slightly over one-fifth of parents (22%) self-reported previously testing positive for HIV (see Table 1).

Table 1.

Dyad sociodemographic and behavioral characteristics

Adolescents (N = 879)	N (%)	Parents (N = 879)	N (%)
Age		Age (years), median	36
14 years	468 (53%)	Age (years), range	18–75
15 years	236 (27%)	Age (years), interquartile range	27–42
16 years	175 (20%)	Gender
Gender		Female	784 (90%)
Female	463 (53%)	Male	89 (10%)
Male	387 (44%)	Transsexual or intersex	1 (0.1%)
Transsexual or intersex	29 (3%)
		Black African	869 (99%)
Black African	822 (94%)
		Primarily IsiXhosa speaking	831 (95%)
		Completed secondary school or equivalent	358 (41%)
		Completed secondary school or equivalent	358 (41%)
Currently attending school	824 (95%)
		Reported household food insecurityb	684 (79%)
Reported individual food insecuritya	264 (30%)
		Marial status
Lifetime sexual partners		Married or cohabitating with partner	274 (32%)
None	480 (55%)	Never married	501 (58%)
1–2 partners	238 (27%)	Widowed, separated, or divorced	86 (10%)
3 + partners	157 (18%)
		HIV status
Ever used a condom		Has tested HIV negative	520 (66%)
Yes	278 (73%)	Has tested HIV positive	176 (22%)
No	104 (27%)	Don’t know or haven’t tested	91 (12%)

^a“Do you ever eat less than you should because there is not enough money for food?”

^b“Did your household ever run out of money to buy food?”

8 adolescents refused to answer school attendance question, 8 refused food security question, 4 were missing number of lifetime sexual partners, and 13 refused condom use question; 5 parents were missing gender, 4 refused to answer education question, 18 refused food security question, 18 refused marital status question, and 92 were missing or refused HIV status question

Among adolescents, 27% (224/833) had heard about PrEP, 67% (497/744) reported that they would “definitely” or “probably” want to use PrEP, and 58% (432/749) thought that their parent would “definitely” or “probably” want them to use PrEP. PrEP interest and perceived parental support were higher among adolescents who had heard about PrEP previously (79% (169/215) and 66% (139/212), respectively) than adolescents who had never heard about PrEP (62% (321/520) and 55% (287/525), respectively). Among all adolescents, those who thought that their parent would support their PrEP use were over twice as likely to be interested in PrEP for themselves (aPR = 2.11, 95% CI 1.82, 2.44, p < .001) (see Table 2). Among parents, 33% (280/853) had heard about PrEP and 85% (558/653) reported that they would “definitely” or “probably” want their child to use PrEP. Parental support for adolescent PrEP use was higher among parents who had ever heard about PrEP (88%; 237/270) than among parents who had not (84%; 317/378).

Table 2.

Log-binomial regression analysis for the association between adolescent perceived parental support and PrEP interest

Adolescent	I would definitely or probably want to use PrEP	I would definitely or probably not want to use PrEP or unsure	Total	Prevalence of PrEP interest	PR (95% CI) p-value	aPRa (95% CI) p-value	Multiply-imputed aPR, (95% CI), p-value
Total N	483	237	720		720	710	879
I think my parent would definitely or probably want me to use PREP	362	59	421	86.0%	2.12 (1.84, 2.45) p < .001	2.11 (1.82, 2.44) p < .001	1.75 (1.46, 2.10) p < .001
I think my parent would definitely or probably not want me to use PrEP or unsure	121	178	299	40.5%	1.00 (ref)	1.00 (ref)	1.00 (ref)

^aadjusted for prior knowledge of PrEP, gender, and lifetime history of vaginal or anal sex

PR = prevalence ratio; aPR = adjusted prevalence ratio; 135 adolescents (15%) refused PrEP interest question and 130 (15%) refused perceived parental support question

Certain PrEP stigma measure items were more commonly anticipated by both adolescents and parents in their respective surveys (see Table 3). For example, 41% of adolescents and 15% of parents agreed or strongly agreed that they “would worry people would think [I/my child] had HIV if they saw [me/him/her] taking PrEP.” Further, 34% of adolescents and 13% of parents thought, “people would judge me if [I/my child] was on PrEP.” Last, 32% of adolescents and 15% of parents thought, “people would give me a hard time if I told them [I/my child] was on PrEP.” Adolescents reported higher average PrEP stigma scores (n = 677, median = 2.43, IQR = 1.86–3.00) than their parents (n = 631, median = 2.00, IQR = 1.00–2.43). Median average PrEP stigma scores were lower among those who had previously heard about PrEP (adolescent n = 194, median = 2.00, IQR = 1.43–2.71; parent n = 266, median = 1.86, IQR = 1.00–2.43) compared to those who had not previously heard about PrEP (adolescent n = 475, median = 2.57, IQR = 2.00–3.14; parent n = 360, median = 2.00, IQR = 1.00–2.43).

Table 3.

Agreement with PrEP stigma measure items among adolescents and parents

Stigma measure item	Adolescents (N = 879) Agree or strongly agree	Parents (N = 879) Agree or strongly agree
	n / total non-missing (%)	n / total non-missing (%)
1) I would feel ashamed of using PrEP / I would feel ashamed if my child was using PrEP	165/757 (22%)	57/652 (9%)
2) I would feel embarrassed about using PrEP / I would feel embarrassed if my child was using PrEP	154/764 (20%)	65/646 (10%)
3) I would think I am not following the ‘rules’ of my community if I took PrEP to prevent HIV / I would think I am not following the ‘rules’ of my community if my child took PrEP to prevent HIV	114/757 (15%)	60/654 (9%)
4) I think people would give me a hard time if I told them I was on PrEP / I would think people would give me a hard time if I told them my child was on PrEP	242/757 (32%)	97/650 (15%)
5) I think people would judge me if I was on PrEP / I would think people would judge me if my child was on PrEP	261/770 (34%)	86/652 (13%)
6) I think I would be at greater risk for physical violence or rape if I was taking PrEP / I would think my child would be at greater risk for physical violence or rape if he/ she was taking PrEP	135/749 (18%)	61/648 (9%)
7) I would worry people would think I had HIV if they saw me taking PrEP / I would worry people would think my child had HIV if they saw him/her taking PrEP	311/764 (41%)	97/655 (15%)

Percentages exclude declinations/missing data

Among all adolescents, those who reported higher PrEP stigma scores were less likely to be interested in PrEP than adolescents who reported lower PrEP stigma scores (aPR = 0.81, 95% CI 0.72, 0.91, p < .001). Similarly, among all parents, those who reported higher PrEP stigma scores were less likely to be interested in their child taking PrEP than parents who reported lower PrEP stigma scores (aPR = 0.83, 95% CI 0.78, 0.89, p < .001) (see Table 4). All multiply imputed models revealed similar statistically meaningful associations as the complete-case analysis.

Table 4.

Log-binomial regression analysis for the association between stigma and PrEP interest/support

Adolescent	I would definitely or probably want to use PrEP	I would definitely or probably not want to use PrEP or unsure	Total	Prevalence of PrEP interest	PR (95% CI) p-value	aPRa (95% CI) p-value	Multiply-imputed aPR, (95% CI), p-value
Total N	440	209	649		649	640	879
Higher PrEP stigma (score ≥ 2.43)	180	122	302	59.6%	0.80 (0.71, 0.89) p < .001	0.81 (0.72, 0.91) p < .001	0.86 (0.77, 0.96) p = .006
Lower PrEP stigma (score < 2.43)	260	87	347	74.7%	1.00 (ref)	1.00 (ref)	1.00 (ref)
Parent	I would definitely or probably want my child to use PrEP	I would definitely or probably not want my child to use PrEP or unsure	Total	Prevalence of PrEP interest	PR (95% CI) p-value	aPRb (95% CI) p-value	Multiply-imputed aPR, (95% CI), p-value
Total N	541	82	623		623	544	879
Higher PrEP stigma (score ≥ 2.0)	244	66	310	78.7%	0.83 (0.78, 0.88) p < .001	0.83 (0.78, 0.89) p < .001	0.88 (0.83, 0.94) p < .001
Lower PrEP stigma (score < 2.0)	297	16	313	94.9%	1.00 (ref)	1.00 (ref)	1.00 (ref)

^aadjusted for prior knowledge of PrEP, gender, and lifetime history of vaginal or anal sex

^badjusted for prior knowledge of PrEP, gender, and self-reported HIV status

PR = prevalence ratio; aPR = adjusted prevalence ratio; PrEP stigma scores were dichotomized at the sample median; 135 adolescents (15%) refused PrEP interest question and 202 (23%) refused at least one PrEP stigma measure item; 226 parents (26%) refused support for adolescent PrEP use question and 248 (28%) refused at least one PrEP stigma measure item

Discussion

In our sample of 879 adolescents 14–16 years of age and their parents or caregivers, less than one-third of participants had previously heard about oral PrEP. Despite low PrEP awareness, two-thirds of adolescents were interested in using PrEP, and although only 58% of adolescents thought their parents would be supportive of their PrEP use, 85% of parents reported they would want their child to use PrEP. Importantly, adolescent perceptions of parental support were associated with their own PrEP interest. This finding is consistent with previous research among adolescent MSM in the United States, which found that perceptions of low parental support contributed to reduced interest in PrEP [8]. However, in our study, parents were more supportive of their adolescent taking PrEP than adolescents perceived they would be. This disconnect may have important implications for PrEP uptake among adolescents. Interventions should aim to improve adolescent-parent communication around sexual health and effective HIV prevention tools.

Anticipated PrEP stigma measure items were endorsed by up to 41% of adolescents and 15% of parents, and adolescents and parents who reported higher PrEP stigma were less likely to be interested in PrEP for themselves and their child, respectively. In previous qualitative research among South African adolescent girls and young women (AGYW) taking daily oral PrEP, AGYW reported experiences of stigma related to PrEP when PrEP was mistaken for HIV treatment and when PrEP was thought to promote sexual promiscuity [9]. In our study, these stigmas were highly anticipated among adolescents in the PrEP stigma measure, suggesting that these factors may impact a range of outcomes across the PrEP continuum. Research is needed, however, to develop a validated measure of PrEP stigma for this context.

Given concerns surrounding PrEP stigma and the low levels of PrEP awareness observed in our study, careful stigma-reducing messaging should be used to increase PrEP knowledge and harness support for PrEP among potential adolescent PrEP users and their parents or caregivers. Research suggests that narratives should focus on protection or wellness framed messages, rather than risk reduction messages, and convey PrEP as an inclusive strategy for anyone in need [12, 24, 25]. Further, in settings with generalized HIV epidemics, the provision of PrEP should be integrated into all relevant health services (e.g., adolescent-friendly clinics, sexual and reproductive health services, primary care, etc.) to help normalize its use and promote PrEP as a component of routine preventative care [24, 25]. Research exploring the development, implementation, and evaluation of interventions that aim to directly combat PrEP stigma among vulnerable populations, including adolescents, is still urgently needed.

We note that there is no validated measure of PrEP stigma. PrEP stigma measure items for this study were created using prior PrEP acceptability research in the South African context [12, 18]. Further, although parents reported lower PrEP stigma scores than adolescents, this may be because parents were assessing the stigma they would anticipate experiencing if their child was taking PrEP. Anticipations of PrEP stigma among parents might differ if they were assessing more proximal stigma associated with taking PrEP themselves. Next, since PrEP was not available to participants in this study, we were not able to assess uptake. Another limitation of this research is that missing data were common, requiring imputation of exposure and outcome values for analysis. Reassuringly, our multiple imputation analysis yielded similar results to our complete-case analysis. Next, we did not collect data on the relationship of the caregivers to the adolescents, which may have helped contextualiz our findings. Lastly, our sample included adolescents with moderate depressive symptoms recruited from a single community. Findings may not generalize to other adolescents with varying risk factors in SSA or more broadly.

This is among the first investigations of parental support for adolescent oral PrEP use. The Our Family Our Future study represents an ideal population of male and female adolescents at an elevated risk for acquiring HIV and who are approaching an age where they might benefit from PrEP. Adolescents are an extremely important group to target with effective biomedical HIV prevention strategies. Adolescents 10–19 years account for 11% of new adult HIV infections globally, with the majority (82%) occurring in SSA [26]. There are important social and behavioral considerations when implementing PrEP use among adolescents, including understanding relationship dynamics. Research among African adults has focused on their ability to harness social support for PrEP from primary partners [27, 28]. However, because adolescents are less likely to be in stable partnerships and more likely to be living with parents [15], parental support may be more important at this age. Future intervention research should explore ways parents can act as allies in supporting adolescents interested in PrEP.

Conclusion

In this sample of adolescents aged 14–16 years and their parents in South Africa, overall PrEP awareness was low, but interest among adolescents and support among parents were high. Parents were more supportive of their adolescents using PrEP than adolescents perceived they would be, and low perceptions of parental support were associated with reduced PrEP interest among adolescents. Further, greater PrEP stigma was associated with lower PrEP interest among adolescents and lower support for adolescent PrEP use among parents. Future research and clinical practice should explore ways parents can act as allies in supporting adolescents interested in PrEP. Interventions to improve PrEP awareness and reduce PrEP stigma are needed for both groups to facilitate adolescent PrEP uptake and use.

Data Availability

Data may contain identifying or sensitive patient information. To preserve participant confidentiality, these data cannot be shared publicly. The study PI, Caroline Kuo (caroline_kuo@brown.edu), may be contacted with requests to access these data for research purposes.