Disclosure of Parental HIV Status to Children: Experiences of Adults Receiving Antiretroviral Treatment at an Urban Clinic in Kampala, Uganda

Limited data are available on the experiences of parental HIV disclosure to children in Uganda. We conducted a qualitative study comprising sixteen in-depth interviews and four focus group discussions with parents receiving highly active antiretroviral therapy. Analysis was done using Atlas.ti qualitative research software. Back-and-forth triangulation was done between transcripts of the in-depth interviews and focus group discussions, and themes and subthemes were developed. Barriers to parents' disclosure included perceptions that children are too young to understand what HIV infection means and fears of secondary disclosure by the children. Immediate outcomes of disclosure included children getting scared and crying, although such instances often gave way to more enduring positive experiences for the parents, such as support in adherence to medical care, help in household chores, and a decrease in financial demands from the children. Country-specific interventions are needed to improve the process of parental HIV disclosure to children and this should encompass preparation on how to deal with the immediate psychological challenges associated with the parent's disclosure.


Introduction
HIV and AIDS continue to be a major public health challenge worldwide more so in sub-Saharan Africa where HIV-related illness is a leading cause of morbidity and mortality. Global estimates indicated that, by the end of 2014, 36.9 million people were living with HIV infection, and of these 25.8 million were in sub-Saharan Africa [1]. Even though there has been a global decline in new HIV infections and associated deaths, the number of clients enrolled for antiretroviral therapy has steadily grown to reach at least 15.8 million [1]. According to a national survey of 2011, about 7.3% of individuals aged 15-49 years are infected with HIV [2]; the number of new infections reportedly dropped from 128068 in 2009 to about 99000 in 2014, while the estimated number of annual HIV-related deaths declined from 52799 in 2009 to 32890 in 2015 [3]. A decrease in HIV-related deaths [4] coupled with an increase in the number of people accessing antiretroviral treatment has given birth to the phenomenon of social normalization of HIV and AIDS [5] which according to Roy and colleagues depends on both disclosure of HIV status to a person's social networks and willingness and ability of the network to support the emotional and material needs of the person living with HIV [6]. In work conducted in Uganda and Tanzania, high adherence resulted from social and material support from an individual's social networks [7].

AIDS Research and Treatment
HIV status disclosure has been the subject of numerous studies on the African continent, particularly in view of increasing incidence of HIV transmission to sexual partners in several countries [8][9][10][11][12][13]. Also, following publication of the World Health Organization (WHO) guidelines on HIV counselling for children up to 12 years of age [14], there has been renewed interest in encouraging parents and caregivers to disclose their HIV status to children. A parent's disclosure of their HIV-positive status can occur at different levels. It can be limited to sharing of health information between the parent and child which then happens to result in the child knowing the parent's serostatus. Disclosure can also be planned by the parent and shared either directly or indirectly: it can take the form of full disclosure when the terms HIV and AIDS are used in sharing the health information, resulting in the child knowing the parent's HIV-positive status; or it can take the form of partial disclosure when the specific terms HIV or AIDS are not used by the parent, yet the parent's actions or conversation results in the child becoming aware of the parent's HIV-positive status.
Available literature indicates that HIV-positive individuals struggle with what, when, and how to disclose their HIVpositive status [15][16][17]. Nondisclosure has been justified by parents based on the premise of protecting children from the negative social and psychological consequences caused by disclosure [16,[18][19][20]. Nevertheless, nondisclosure has been associated with negative consequences for both the parent and child [21,22]. In many developing countries, there is a growing demand for culturally appropriate interventions intended to increase parents' readiness to disclose their HIV status to their children [15]. Interventions to encourage parental HIV disclosure to children have been tested in the United States [21,23] and China [24]. However, in sub-Saharan Africa, with the exception of South Africa [25][26][27], such interventions are uncommon. Moreover, the implementation of these kinds of interventions should be grounded on a good understanding of the local context regarding potential parental HIV disclosure. In Uganda, limited data are available on the barriers, approaches, motivators, and consequences of parental disclosure of HIV status to children. This scarcity of information provided the impetus for this study.

Study Design and Setting.
A cross-sectional, descriptive, qualitative study was conducted among HIV-positive clients in care and treatment at the Infectious Diseases Institute at Makerere University's College of Health Sciences, Kampala, Uganda. The Infectious Diseases Institute is a Ugandan notfor-profit organization, established in 2002, whose mission is to strengthen HIV-related care, training, and research in Africa. By 2014 the Infectious Diseases Institute was providing care and treatment services to over 110000 people living with HIV in urban and rural settings in Uganda, which amounts to about 13% of the national effort [18].

Participant Recruitment and Data
Collection. The study participants comprised adult parents who were HIV-positive, who either had or had not disclosed their HIV serostatus to their child/children, and who had a biological child aged at least 5 years or older. The age-5 lower limit was chosen based on a recommendation by the World Health Organization [14] that children of school age should be told of a caregiver's HIVpositive status. No upper limit cut-off for a child's age was stated because in African settings parents tend to perceive children as a child irrespective of their child's age. The sample was purposively determined to include both clients who had disclosed their HIV status and those who had not. Potential participants were identified by the research assistants at the clinic; to avoid disrupting the smooth running of the clinic, willing clients were given appointments to return later for an in-depth interview (IDI) and/or focus group discussion (FGD). The selection involved a process of screening all potential participants during the time when they were waiting for review of their individual treatment. Each client who met the inclusion criteria was then given an appointment for when to return for the purposes of participating in the study. Sampling was also guided by the principle of data saturation-that is, data collection was terminated when no new information appeared to be elicited from the IDIs and FGDs.
Data were first collected from participants who returned for an IDI. Each parent was taken through the process of consent, and those who provided written informed consent were then individually interviewed by one of four trained research assistants. Each interview lasted between thirty and forty-five minutes and all the interviews were conducted in Luganda (the local language of central Uganda). To ensure privacy and confidentiality, each interview was conducted in a quiet room usually used for counselling clinic clients. Each of the participants in the study, in either an IDI or FGD, was paid the equivalent of US $10 to compensate for travel and time. A total of 16 IDIs were carried out with HIV-positive parents who had or had not disclosed their status to their child/children. Of these, nine parents had disclosed their HIV status (i.e., shared health information with a child, resulting in the child knowing the parent's HIV-positive status) and seven had not. In addition, four FGDs, each with 8 or 9 participants, were carried out: two groups with men and two groups with women. We chose to conduct separate FGDs for men and women for cultural reasons, as in the Ugandan context anecdotal evidence suggests that women tend to express their opinions less freely in the presence of men. We conducted one FGD for men and one for women who had disclosed their HIV status, and likewise there was one group for men and another for women who had not disclosed their HIV status. A total of 35 clients participated in the four FGDs. The IDIs and FGDs were conducted using guides developed by the investigators after reviewing the literature on parental HIV disclosure. Different guides were developed for research with parents who had disclosed their HIV status and with those who had not. All the guides were translated into Luganda by an expert from Makerere University's Department of Linguistics, English-Language Studies and Communication. The translated guides were pretested among HIV-positive parents attending the Old Mulago HID and AIDS Clinic in the general outpatient department, which was not part of the study site so as to avoid bias that would result from patients AIDS Research and Treatment 3 who participated in the pretest being included in the study. Adjustments were made to the flow of questions and probes before the data collection began.

Data Management.
Audio-recordings of the IDIs and FGDs were transcribed and translated and then analysed by an expert with a social science background, using Atlas.ti qualitative data management software. The transcripts were assigned a name that did not contain any personal identifying information, to maintain anonymity. A code book was developed after reading through a random sample of five transcripts (2 FGDs and 3 IDIs) to ascertain the emerging themes. The transcripts were then entered into Atlas.ti. Next, the codebook from the emerging issues that had been transformed into different codes was entered into Atlas.ti. Each of the transcripts was then thoroughly read and any portion of text identified that related to a particular code was highlighted and attached to the respective code. Query reports for each of the codes were produced containing short summaries of texts. Each of these short summaries was read several times and analysed manually using manifest content analysis to identify emerging patterns. Back-andforth triangulation was done between transcripts of the FGDs and IDIs before a decision was made to group those sources together under the themes and subthemes. These themes reflected underlying meanings in the text. In presenting the results here, we quote extractions from the texts, representing the themes, to support our findings.

Sociodemographics of the Participants.
A total of 51 parents on ART participated in the study. Sixteen clients participated in IDIs: 7 males and 9 females; their ages ranged from 28 to 59 years. Thirty-five individuals participated in FGDs: 19 males and 16 females; their ages ranged from 24 to 67 years. Religious affiliations and disclosure status of each participant are shown in Table 1.

Parents' Experiences regarding HIV Disclosure.
Experiences concerning parental HIV disclosure fell into four categories: barriers/hindrances, motivators, approaches, and consequences. Themes that emerged from the data in each category are summarised in Table 2. Table 1 summarises four themes that emerged from the data relating to barriers or hindrances to parents' disclosure: viewing disclosure as a burden, perceived immaturity of the children, strength of the relationship between the parent and child, and anticipated unfavourable level of support from the child.

Viewing Disclosure as an Unnecessary
Burden. Some parents who had not disclosed their HIV status viewed sharing their serostatus as an unnecessary burden that was not worth the risk of going through it. They feared that their child/children might ask about their own HIV status, that they would probably tell the other partner/parent about their HIV test result, and that this would exert pressure on the parent to explain how the disease came into the family. One male participant in a FGD said the following: 3.3.2. Children Perceived as Too Young to Understand. Some parents chose not to disclose their HIV status because they perceived their child as too young to understand the meaning and consequences of HIV infection, and some also believed a child would not be able to keep the parent's HIV status confidential. One woman explained the following:

I take drugs with my husband, and [the children] donot see it, we keep it in the bedroom. Even if they go there they
have not yet known it, they have never seen it anywhere. They will only see the tin, they will not know whether it is for headache or stomach pains. I will tell them when they grow up, but now they are very young, even if you tell them they will not take it seriously. (IDI with female)

Poor Relationship between Parent and Child.
Parents, but particularly fathers, often chose not to disclose their HIV status to their child because they felt the relationship between them was not good enough to facilitate disclosure. Some parents reported lacking the personal confidence to disclose their HIV status, and in other instances they felt their child was not interested in their health. Other reasons included parents viewing their adult children as being interested only in sharing the parent's property and less so in personal concerns. There were also cases where the children were regarded as being close to only one parent, which made the other uncomfortable about disclosure. This is demonstrated in explanations provided by two men:   Table 2: Barriers/hindrances to disclosure as presented by parents who did not disclose HIV serostatus to their children.

Themes Subthemes
Viewing disclosure as a risk or burden (i) Parent being in poor health and therefore concerned more about their health than disclosure.
(ii) Parents perceived the children as already knowing their (parent's) HIV status.
(iii) Parents feared that the children may ask them about their own (child's) HIV status.
(iv) Parents feared their children's assumption that they (parents) are about to die.
(v) Parents preferring to have "peace of mind" and hence avoid disclosure to children.
Immaturity of children Anticipated support from children (i) Parent viewed children as not being able to support them emotionally or practically.
(ii) Parent still felt in good health and saw no need to be supported by the children.

Anticipated Unfavourable Level of Support from a Child.
Some parents felt self-reliant because they were in otherwise good health and able to support themselves and thus did not feel the need or benefit of disclosure to their children. One man explained the following:  (Table 3).
Overall, men were more likely to be motivated to disclose their HIV status to their child so as to clarify doubts about their HIV-positive status, whereas females were more likely to disclose their HIV status as a result of seeing it as an obligation to their child.   (Table 5) included themes wherein a child showed concern about the parent's health, parents received support from the children, children improved their behaviour, and the parent felt psychological relief after disclosure.

Children Showing Concern about a Parent's Health.
In some cases, once a parent had disclosed their HIV-positive Table 4: Some processes of disclosure, as presented by the patients that had disclosed their HIV status to their children.

Themes Subthemes
Being prompted by events (i) When the children asked about the parent's HIV status.
(ii) When the children wanted to know their own HIV status.
Gathering children (i) Parents planning well in advance and gathering the children together for the disclosure.
Offering clues (i) Leaving the antiretroviral medicines for the children to see, without telling them that the parent is HIV-positive. Table 5: Consequences of disclosure, as presented by parents that had disclosed their HIV status.

(a) Positive consequences
Support and concern about health of the parent (i) Reminding parents when the time for taking their drugs approached.
(ii) Reminding their parents about the next hospital appointment.
(iii) Financial assistance or helping with household chores.
Improved behaviour of children (i) Change in the behaviour of the child to avoid stressing the parent.

(b) Negative consequences (mainly short term)
Emotional breakdown (i) Children responding to disclosure by crying.
(ii) The parent themselves crying after disclosure.
(iii) Stress for both parent and child. Children getting scared (i) Children responding by getting very scared.
status, a child started reminding them to take their medicines and to keep hospital appointments. One father said the following: And when time comes they tell me that "Father, your medicine?" And sometimes I take like a week when I am away from home, but they will ask you that "Daddy, how many days will you spend there?" I tell them that may be one week, and they will pack medicines for two weeks, and they will tell you that "I have put for two weeks because you might reach there and anything can happen." (IDI with male)

Parents Receiving Emotional or Practical Support from
Children after Disclosure. Some children had responded to their parent's disclosure by offering emotional, material, financial, or even food support to the parent. Their support also included help with household chores: Since the children got to know about my status they provided a lot of care because they know their mum is sick; they do most of the work at home, especially the boy does not want me to go and fetch water and he fetches it himself. . .so that is it. . .they help me a lot because they do all the work. (IDI with female) 3.6.3. Improved Behaviour of Children. There were cases when children were described as making excessive financial demands before the parent had disclosed their HIV status, yet after disclosure the parent reported a positive change in a child's behaviour and this lessened the parent's stress. One woman said the following: So they understood and they felt concerned, and they started to give me support and they reduced on the demands they were putting on me; those who used to ask money for special hire to take them to school started taking a taxi to go to school. (IDI with female) 3.6.4. Parents' Psychological Relief. Some parents reported a sense of relief after disclosure to their children. One mother said the following: Those other ones were also hurt, but one of the children was more hurt, she showed that it was too much for her.

Discussion
As supported by our findings, parents' HIV status disclosure to children is still a challenging issue characterised by numerous obstacles, varied approaches, and mixed outcomes. We found it interesting that parents who perceived themselves as self-reliant and in otherwise good health did not consider it essential to disclose their HIV-positive status to their children. In addition, however, we observed many of the usual other hurdles to disclosure, as previously described in the literature, such as parents' perceptions that children are too young to understand the health information and that they will be subsequently unable to keep it secret [16,18,[28][29][30], fear of having to explain how a parent became infected [31], and perceptions of a poor relationship between the parent and child [29]. In this study, parents' motivation for disclosure sometimes clearly related to their improved health status following an episode of severe illness. To our knowledge, this finding has not been reported by other studies in Uganda. Parents had also often disclosed their HIV status out of a sense of responsibility and obligation to their children. Similar findings have been reported by studies conducted in the United States [32] and China [33]. Importantly, this finding provides an opportunity for healthcare professionals to build on parents' inner strength and desire to protect, beyond just making the simple recommendation that parents should disclose their HIV status to their children. Thus, future interventions could aim to bolster parents' propensity to discharge their perceived responsibility in this regard. Moreover, parents who are provided with basic information and skills about disclosure are likely to be encouraged and better equipped to disclose their HIV status to their children, and this can include the benefits of improved family communication and cohesion [34,35].
Some parents in this study cited the need to protect adolescent children from engaging in higher-risk sexual behaviour as a strong consideration for their disclosure. This finding supports the results of studies conducted in China [33], Burkina Faso [29], and the United States [35]. One implication is that the promotion of HIV status disclosure, particularly in the context of family, may pay significant dividends in regard to HIV prevention. This is so because parents who discuss with children measures to prevent HIV infection are more likely to be trusted and believed than other adults because of their experience and position in the family. This supports the need for approaches that enhance parental self-efficacy in HIV disclosure. Whereas in this study disclosure was often portrayed as a protective action, the literature is rife with examples where nondisclosure has been justified by parents based on the same premise of protecting children from the negative social and psychological consequences caused by disclosure [16,[18][19][20]. Nevertheless, while nondisclosure is sometimes intended to protect children, it has been associated with negative consequences for both the parent and child [21,22]. Additional research, especially in resource limited countries with a high burden of HIV, is merited to demonstrate whether the benefits of parental disclosure far outweigh the negative outcomes.
In this study, the process of disclosure for some parents was hastened by circumstances, such as children asking parents how long they (child/children) were going to be on a medication. For other parents, disclosure was a preplanned process that involved gathering their children together for a family discussion which culminated in the parent sharing his or her HIV-positive status. Similarly, Demattero and colleagues (2002) reported that disclosure was sometimes planned in the context of gathering for a special meal or favourite family activity. The same authors stated that disclosures had also occurred in the context of a family argument or after a parent had been drinking. The experiences of parents in this study and others reveal that there is a dearth of practical and generalizable approaches to help parents in making decisions regarding the process of disclosure. Future work in this area, especially in Uganda and similar African contexts, should focus on developing contextually appropriate interventions to encourage parents' HIV status disclosure to children.
In this study, the outcomes of parental HIV disclosure were mixed, as has been similarly demonstrated in other studies. Positive outcomes noted by parents after disclosure included children reminding the parent about taking their medication, packing medicines for the parent when they had to travel, helping more with household chores, and improving their behaviour, including reducing the financial demands the child put on the parent; comparable findings were reported by Kennedy et al. (2010) and Tiendrebeogo et al. [29]. Conversely, negative consequences of disclosure tended to be the immediate emotional burden, as many of the parents reported that a child had reacted by crying or getting scared. Perhaps it is primarily for fear of these emotional reactions that many parents avoid disclosure to children [36]. Yet, despite the negative consequences associated with parental HIV disclosure, it is suggested that parents' nondisclosure to children might actually lead to increased psychosocial vulnerability [22].

Study Limitations
The findings of this qualitative study cannot be generalized to all HIV-positive parents in Uganda. Furthermore, when interviews are used as a method of data collection, there is a tendency of participants to distort their responses to present a favourable opinion, which leads to response bias. This was minimized by emphasizing honesty throughout the data-collection period. Finally, it would have been interesting to document the consequences of parental HIV disclosure from the perspectives children, but this was outside the scope of this study. Despite these limitations, the study presents important findings that can help shape the direction of interventions to promote parental HIV disclosure in Uganda. Approaches to facilitate parental HIV disclosure in Uganda and other highly affected resource limited settings are especially needed following the recently released World Health Organization policy guidelines requiring all individuals who test HIV-positive to commence ART [37]. These policy guidelines have resulted in development of country-specific "test and treat" guidelines thus transforming HIV into a chronic infection.

Conclusions
Parental HIV disclosure remains a complex phenomenon hampered by psychological barriers and uncertain family dynamics. In situations where disclosure occurs, the outcomes may be mixed for both the parent and child. However, HIV-treatment programmes in sub-Saharan Africa require culturally suitable interventions that facilitate the process of disclosure. Particularly, interventions are needed to better prepare parents and children to deal especially with the immediate psychological problems associated with parental HIV status disclosure. In the meantime we recommend that parents and children should receive counselling before and after disclosure to deal with psychological challenges associated with disclosure.

Consent
Written informed consent was obtained from each participant before data collection.