AIDS and Behavior

, Volume 15, Issue 6, pp 1121–1127

Prevalence and Pattern of Disclosure of HIV Status in HIV-Infected Children in Ghana

Authors

  • Stacey Kallem
    • Department of Pediatrics and PharmacologyYale University School of Medicine
  • Lorna Renner
    • Department of Child HealthUniversity of Ghana Medical School
  • Musie Ghebremichael
    • Department of BiostatisticsHarvard University
    • Department of Pediatrics and PharmacologyYale University School of Medicine
Original Paper

DOI: 10.1007/s10461-010-9741-9

Cite this article as:
Kallem, S., Renner, L., Ghebremichael, M. et al. AIDS Behav (2011) 15: 1121. doi:10.1007/s10461-010-9741-9
  • 328 Views

Abstract

With the advent of highly active antiretroviral therapy (HAART) HIV-infected children are surviving into adulthood. Despite emerging evidence of the benefits of disclosure, when and how to disclose the diagnosis of HIV to children remain a clinical dilemma. We investigated the prevalence and determinants of HIV disclosure in a cross-sectional study of 71 caregiver-child dyads from the Pediatric HIV/AIDS Care Program at Korle-Bu Teaching Hospital (Accra, Ghana). The children were between 8 and 14 years of age (median age, 10.39 years). The prevalence of disclosure was 21%. In the unadjusted analyses, the age of child, the level of education of child, deceased biologic father, administration of own HIV medications, and longer duration on HIV medication were significantly associated with disclosure. The low prevalence of disclosure underscores the need for a systematic and a staged approach in disclosing HIV status to infected children in resource limited countries.

Keywords

Pediatric HIVDisclosureCaregiver-child dyadResource limited countries

Introduction

Children under 15 years of age account for approximately 2 million of the people living with HIV worldwide. Almost 90% of all HIV infected children live in sub-Saharan Africa [1]. With the advent of antiretroviral therapy (ART), there has been a significant reduction in morbidity and mortality of HIV-infected children and more of them are surviving through childhood and into adolescence. Along with the increased survival of children infected with HIV, disclosure of HIV status to children remains a complex and a critical clinical issue in the care of HIV-infected children. Caregivers may be reluctant to disclose the HIV status to their children for fear of social rejection and isolation, parental sense of guilt, and fear that the children would not keep their diagnoses to themselves [2]. However, several studies have documented the benefits of disclosure of the HIV status to HIV-infected children, including psychological benefits as well as positive effects on the clinical course of the disease [35].

The American Academy of Pediatrics recommends that children and adolescents with HIV should be told their diagnosis [6]. However, published rates of disclosure in children from resource rich countries vary widely, from 18 to 77% [2, 69], partly due to the lack of conclusive guidelines on when and how to disclose the diagnosis of HIV to children. Though limited, studies available on rates and effects of HIV disclosure on children are mainly from resource rich countries. There is paucity of data on disclosure from resource limited settings, where the majority of HIV infected children live. Moreover, studies from resource limited settings have mainly been qualitative in design with small samples sizes making it difficult to generalize the findings. In the present study, we sought to address some of the limitations of previous studies on disclosure from resource limited settings. We investigated the prevalence, pattern, and predictors of HIV disclosure in HIV-infected children in one of the tertiary hospitals in Ghana.

Methods

Setting

The Pediatric HIV/AIDS Care program at Korle-Bu Teaching Hospital, Accra, Ghana was established in 2004 and provides comprehensive HIV/AIDS care and management of opportunistic infections. The children referred to the clinic either have mothers already known to be HIV seropositive during pregnancy through the program for the prevention of mother-to-child transmission (PMTCT), are discovered to be infected with HIV after presenting with an AIDS-defining illness, or are diagnosed after either a symptomatic sibling or parent was found to be HIV positive. The majority of the children were infected perinatally. The program takes care of about 700 children including those with confirmed HIV diagnosis and HIV-exposed children pending final status determination.

Participating families were identified through the Pediatric HIV/AIDS Care program. All the children recruited into the study received medical care for their HIV infection through the clinic. Over the period of the study, June to August, 2009, all the families attending the clinic with children aged 8–14 years were offered participation in the study by their primary medical provider. All the children enrolled into the study had been diagnosed with HIV infection. In addition, nine healthcare providers working at the pediatric HIV clinic were enrolled into the study.

All participants gave informed consent in order to participate and child participants required both parental/caregiver consent and child assent. Participants were given the opportunity to refuse or withdraw from the study at any point. All participants were given a small stipend to offset travel expenses. The study was approved by the Institutional Review Boards of the University of Ghana Medical School and the Yale University School of Medicine, USA.

Study Design

A cross-sectional study of 71 caregiver-child dyads was conducted from June to August, 2009. Participants were administered a study questionnaire, the design of which was informed from previous studies on HIV disclosure [10, 11]. Interviews were conducted by a trained interviewer and lasted about 25 min. The questionnaire was administered mainly in English. In few instances, a trained multi-lingual interpreter assisted in administering the questionnaire in a local language depending on the language proficiency of the study participant. Information was collected on disclosure status, age at disclosure, treatment compliance, and socio-demographic data, family type (biologic parents, adoptive/foster, extended), and caregiver’s HIV status. Children and caregivers were read separate structured questionnaires. To avoid inadvertent disclosure of a child’s status, children were interviewed in the presence of their caregivers while the caregivers were interviewed without the children present. To assess whether the children knew that they were HIV positive, they were first asked an open-ended question of “Can you tell me why you are at the doctor today?” Follow up questions were asked to elicit the specific name of their illness. Caregivers were asked questions such as “do you think your child knows that he/she has HIV/AIDS?”, “when talking to your child about doctor visits and why he/she is sick, do you use the words HIV/AIDS?”, and “why did you choose to disclose or not disclose to your child his/her HIV status?” The primary care providers in the clinic also completed a brief questionnaire comprising questions on who should disclose the child’s HIV status and appropriate age for disclosure.

Medical records were reviewed for medical diagnoses, reported route of transmission, disease severity as determined by WHO clinical staging, current medication including type, treatment compliance, and CD4+ count.

Outcome Variables

The main outcome variables were disclosure and non-disclosure of status. HIV disclosure status was based on the caregiver report. Caregivers were asked whether the child knows his/her HIV status.

Definitions

Disclosure: Where the caregiver said that the child knows his/her HIV diagnosis.

Non-disclosure: Where the caregiver said that the child does not know his/her infection or where the caregiver was unsure if the child knew his/her status.

Caregiver: A person who lives with the child, participates in the child’s daily care and is the most knowledgeable about the child’s health. They were either biological parents or guardians acting as surrogate parents to the child.

Orphan: A child that has lost either one or both parents.

Statistical Analysis

HIV disclosure status was based on whether the caregiver said the child knew his/her diagnosis. Fisher’s exact tests were used to compare categorical variables between children who knew their HIV status and those who did not know. For continuous variables, Wilcoxon rank sum tests were used. Multivariate analyses using logistic regression models were employed to examine the predictors of HIV disclosure. Two sided P-values are reported.

Results

Characteristics of Study Participants

The demographic characteristics of the 71 caregiver-child dyads enrolled into the study are illustrated in Tables 1 and 2. Fifty-four percent of the children interviewed were females and 60% were in a grade level of three or higher. Almost half of the children were orphans. The percent of biological parents with HIV was 80 and 52 for mothers and fathers, respectively. Seventy-five percent of the caregivers had at least completed junior high school. Over 50% of the caregivers had at least a monthly income of 200 GHC (approximately 130 US Dollars). The HIV disease activity of the children has been summarized in Tables 1 and 2. The mean age of the children was 10.42 years (SD = 1.72). The HIV clinic was started in 2004 and the mean duration of attendance at the clinic by study children was 36.25 months (SD = 21.59). At the time of registration into the clinic, 62% of the study children had WHO clinical stage 3 or 4. Eighty-five percent of the children were on HAART at the time of enrollment into the study; the mean duration on HAART was 29.92 months (SD = 18.17). The HAART regimen comprised of zidovudine (AZT), lamivudine (3TC), and efavirenz (EFV). The mean CD4 absolute count and percent values at or closest to the time of study was 856 (SD = 631.93) and 26 (SD = 13), respectively (Table 2).
Table 1

Characteristics of the study population

Characteristics

N

%

Gender of child

 Male

33

46

 Female

38

54

Education of child

 Class 1–2

28

40

 Class 3–4

28

40

 Class 5 or more

14

20

WHO staging of childa

 1

6

8

 2

21

30

 3

36

51

 4

8

11

Mother alive

 No

31

44

 Yes

40

56

Father alive

 No

32

46

 Yes

37

54

ARV status of child

 Not on ARV therapy

11

15

 On ARV therapy

60

85

Caregiver education

 None

8

11

 Elementary school

10

14

 Junior high school

32

46

 Secondary school

20

29

Monthly income of household

 <100 GHC

28

41

 200–300 GHC

27

40

 >300 GHC

13

19

Mothers HIV status

 Negative/unknown

13

20

 Positive

51

80

Fathers HIV status

 Negative/unknown

31

48

 Positive

34

52

Who supervises ARVs

 Child

24

41

 Father/mother

23

40

 Othersb

11

19

ARV antiretroviral, GHC Ghana Cedis (1 GHC is equivalent to 0.65 USD)

aWHO staging: 1, no symptoms; 2, mild symptoms; 3, advanced symptoms; and 4, severe symptoms

bOthers: grandparents, sister, uncle, aunt, stepmother, schoolmother

Table 2

HIV disease activity of study children

Variable

Children

Median (range)

N

Mean (SD)

Age in years

63

10.42 (1.72)

10.42 (8.0–14.23)

Time on ARVs (months)

61

29.92 (18.17)

31.74 (0.92–62.16)

Duration of clinic attendance (months)

70

36.25 (21.59)

37.69 (0.00–106)

CD4 absolute counta

69

856 (631.93)

843 (16–3921)

CD4 percent

61

26.07 (12.95)

29 (1.00–54.50)

ARVs antiretrovirals

aNumber of cells per μl of whole blood

The Prevalence of HIV Disclosure

According to caregiver account, 21% of the children (15 of 71) had their HIV status disclosed to them. Of these 15 children, only 7 (47%) mentioned HIV in describing their own illness; while 8 (53%) did not make reference to HIV in describing their illness.

Caregiver Reasons for Disclosure or Non Disclosure

Caregivers were asked questions about their reasons for choosing to disclose HIV status or not to their children. Of the 15 caregivers who had disclosed to their children, over half (eight) cited the child questioning or refusing to take medications or come to clinic appointments as a reason for disclosure. Additional reasons for disclosure included the insistence of the doctor (four), age of the child (two), and so the child could protect himself/herself (four).

Fifty-six caregivers had not disclosed HIV status to their children. Over half (29) of these caregivers cited fear that the child would tell others about his diagnosis as a reason for non-disclosure. Additionally, 26 of the caregivers believed their children were too young to know their status. Lastly, 14 caregivers chose not to disclose because they thought that if they knew their status, the children would be worried, scared or think that they were dying.

Association Between Demographic Characteristics and HIV Disclosure

Univariate analyses using Fisher’s exact and Wilcoxon rank sum tests were employed to examine the association between the characteristics of the study participants and the HIV disclosure status (Tables 3, 4). The only caregiver characteristic significantly associated with disclosure of HIV status to the child was having a deceased biologic father (Fisher’s exact test, P = 0.02). Whether the biological mother was alive or dead, HIV status of biological parents, education attainment of caregiver, age of caregiver, and household income were not significantly associated with disclosure of HIV status to the child. The characteristics of the child that were significantly associated with disclosure of HIV status were level of education (Fisher’s exact test, P < 0.01), child responsible for taking his/her own medication (Fisher’s exact test, P = 0.02), age of child (Wilcoxon rank sum test, P < 0.01), duration of starting on HAART (Wilcoxon rank sum test, P = 0.02), and duration of clinic attendance (Wilcoxon rank sum test, P = 0.04). The ARV status, WHO clinical staging at clinic registration, and CD4-T lymphocyte count at time of interview were not associated with the disclosure status of the child.
Table 3

Demographic characteristics by disclosure status of child

Characteristics

No

Yes

P-valuea

N = 56

%

N = 15

%

Gender

    

0.38

 Male

28

50

5

33

 

 Female

28

50

10

67

 

Education

    

<0.01

 Class 1–2

26

47

2

13

 

 Class 3–4

24

44

4

27

 

 Class 5 or more

5

9

9

60

 

WHO stagingb

    

0.68

 1

4

7

2

13

 

 2

16

29

5

33

 

 3

30

53

6

40

 

 4

6

11

2

13

 

Mother alive

    

1.00

 No

24

43

7

47

 

 Yes

32

57

8

53

 

Father alive

    

0.02

 No

21

39

11

73

 

 Yes

33

61

4

27

 

ARV status

    

1.00

 Not on ARV therapy

9

16

2

13

 

 On ARV therapy

47

84

13

87

 

Care giver education

    

0.39

 None

7

13

1

7

 

 Elementary school

8

14

2

13

 

 Junior high school

27

49

5

33

 

 Secondary school

13

24

7

47

 

Monthly income of household

    

0.24

 <100 GHC

23

43

5

36

 

 200–300 GHC

23

43

4

28

 

 >300 GHC

8

14

5

36

 

Mothers HIV status

    

1.00

 Negative/unknown

10

20

3

20

 

 Positive

39

80

12

80

 

Fathers HIV status

    

0.25

 Negative/unknown

26

52

5

33

 

 Positive

24

48

10

67

 

Who supervises ARVs

    

0.02

 Child

14

31

10

77

 

 Father/mother

21

47

2

15

 

 Othersc

10

22

1

8

 

ARV antiretroviral, GHC Ghana Cedis (1 GHC is equivalent to 0.65 USD)

aFisher’s exact test P-value

bWHO staging: 1, no symptoms; 2, mild symptoms; 3, advanced symptoms; and 4, severe symptoms

cOthers: grandparents, sister, uncle, aunt, stepmother, schoolmother

Table 4

HIV disease activity by disclosure status of child

Variable

No (N = 56)

Yes (N = 15)

Wilcoxon test statistic

N

Mean (SD)

N

Mean (SD)

Age in years

49

10.05 (1.48)

14

11.72 (1.93)

621**

Time on ARVs (months)

48

27.1 (16.81)

13

40.2 (19.95

531*

Duration of clinic attendance (months)

56

33.7 (19.13)

14

46.3 (28.09)

641*

CD4 absolute counta

54

852 (635.5)

15

871 (640.5)

532

CD4 percent

48

25.4 (13.49)

13

28.6 (10.8)

457

ARVs antiretrovirals

** P < 0.01, * P < 0.5

aNumber of cells per μl of whole blood

Predictors of Disclosure of HIV Status

Multivariate analyses using logistic regression models were employed to examine characteristics of children or caregivers that could predict the HIV disclosure status of a child. All the factors that were associated with HIV disclosure status at the 0.30 level in the univariate analysis were included in the multivariate analysis (Table 5). Age, level of education of the child, deceased biologic father, administration of own HIV medication remained the main predictors of HIV disclosure status in the unadjusted models. The children were more likely to know their HIV status, if they were older, in class 5 or higher, in charge of taking their own ARVs, had a deceased father, had been on ARVs longer, or been a clinic attendee longer. In the adjusted model, none of the above variables predicted the HIV disclosure status of a child, thus, only unadjusted odds ratios are reported in Table 5.
Table 5

Predictors of HIV disclosure status

Predictors

Unadjusted odds ratios

OR (95% CI)

Education

 Class 1–2

1.00

 Class 3–4

3.63 (0.35–37)

 Class 5 or more

92.0** (7.32–999)

Father alive

 No

1.00

 Yes

0.20* (0.05–0.84)

Father HIV

 No

1.00

 Yes

2.21 (0.57–8.54)

Monthly income

 <100 GHC

1.00

 200–300 GHC

1.31 (0.28–6.07)

 >300 GHC

3.00 (0.59–15.3)

Who supervises ARVs

 Child

1.00

 Father/mother

0.07* (0.01–0.58)

 Othersa

0.16 (0.02–1.43)

Age in years

1.86** (1.21–2.86)

Time on ARVS (months)

1.05* (1.01–1.10)

Time on clinic (months)

1.04* (1.00–1.08)

GHC Ghana Cedis (1 GHC is equivalent to 0.65 USD), ARVs antiretrovirals

** P < 0.01, * P < 0.5

aOthers: grandparents, sister, uncle, aunt, stepmother, schoolmother

Interview of Healthcare Providers

We interviewed nine healthcare providers in the clinic for their views on HIV disclosure to their patients. The providers comprised of two attending physicians, two resident physicians, three pharmacists, and two nurse/counselors. Seventy-eight percent of them said disclosure should be done by the caregiver and 18% said it should be by both caregivers and clinic providers. However, when caregivers who had not yet disclosed were asked the same question, nearly a third of them wanted to defer disclosure to the clinic personnel.

All the healthcare providers thought disclosure should be a process with full disclosure by 12–13 years of age. The providers thought formal training in counseling and disclosure was critical.

Discussion

The prevalence of disclosure among our study population was 21%. This is within the lower bounds of the range (18–77%) reported in studies from resource rich countries [2]. Our finding is consistent with reported prevalence of disclosure from Thailand, Zambia, and Uganda of 30.1, 31.8, and 29%, respectively [4, 10, 12]. Several studies from both resource rich and resource limited countries suggest that the disclosure of the HIV status to HIV-infected children has psychological benefits and positive effects on the clinical course of the disease [4, 5, 13]. Disclosure of HIV status helps HIV infected children to understand their illness and the need for treatment; thereby promoting their participation and responsibility for their treatment [14]. The rate of disclosure remains low especially in resource limited countries despite the growing evidence of the benefits of disclosure. Reasons cited by the caregivers were consistent with that of studies in resource-limited countries; namely stigmatization and isolation, parental sense of guilt, and fear that the child would not keep diagnosis to themselves [2]. Our findings and that of other studies in the sub-region of low rate of disclosure underscore the need for a systematic review of the hurdles to disclosure and implementation of locally and culturally-sensitive interventions programs to promote disclosure. The low rates of disclosure in pediatric HIV patients may reverse the gains of the unprecedented global initiative to scale-up ARV access in resource-limited countries.

The inability of most caregivers to handle disclosure has defined the three main patterns of disclosure: complete parental, partial, and non-disclosure. In complete disclosure, the child is told that he/she has HIV and is given disease specific information; while, in partial disclosure, the child may know that he/she has an illness but he/she has not be told specifically that he/she has HIV infection. In complete disclosure, the child is aware and refers to his/her illness as HIV related. Complete disclosure of HIV status has been associated with improved adherence to ART [12]. Partial disclosure and non-disclosure can strain the relationship between the caregiver and the child. Force and persuasion are often used to get the child to adhere to treatment; these may results in purposeful rebellion and non-adherence by the child [12]. In our study, the prevalence of complete and partial disclosures was 47 and 53%, respectively.

Nearly a third of the caregivers expected the healthcare providers alone to disclose the HIV status of their children. This is contrary to a Thai study where majority of caregivers believed that as parents it was their responsibility to disclose the diagnosis to the child [10]. This may be partly due to the cultural differences in relation to communication between parents and children on issues and the lack of disclosure skills. The healthcare providers agreed that disclosure is the prerogative of the caregivers and children should be told their HIV status. These contrasting views are consistent with that of studies from resource rich countries [15]. The expectation of some caregivers that healthcare providers should disclose the HIV status to the HIV-infected child may lead to unnecessary delays in disclosure. As reiterated by the healthcare providers, training of the healthcare providers in disclosure counseling and a disclosure program adopted by the clinic would allow healthcare providers to better advise caregivers on how to disclose. With caregivers who lack disclosure skills, a counselor-assisted or supported disclosure session may suffice [12].

In our study, the factors that were associated with disclosure were the age of the child, the level of education of child, deceased biologic father, administration of own HIV medications, and longer duration on HIV medication. The finding of the association between disclosure and older age, the level of education, and relative’s illness or death are consistent with previous studies [16, 17]. Disclosure has been associated with severity of HIV/AIDS, lack of adherence to treatment, the child’s expressed will to know his/her condition, HIV-seronegative caregiver, caregivers with good support system, family expressiveness, and higher socioeconomic status [2]. In our study, socioeconomic status of caregiver, WHO clinical staging, and CD4-T lymphocyte count at time of interview were not associated with disclosure. Several studies in resource rich countries have reported that parents’ own HIV status can influence both decision-making around disclosure of the child’s status and the disclosure process itself [18]. An interesting finding was the fact that children who were responsible for taking their medication without caregiver supervision were more likely to know their status. Data on adherence was not uniformly available so we could not model adherence in our analyses. Other studies have found that disclosure promotes adherence to treatment and in our study nearly 50% of the caregivers who had disclosed cited issues surrounding medication adherence as a reason. It is possible that after disclosure, children took charge of their illness and medication leading to improved adherence. In studies in Zambia and the Democratic Republic of the Congo, a common reason caregivers gave for disclosure is the hope that by knowing their status, the children will have better adherence to treatment [4, 11].

In a two-level mixed model, age, level of education of the child, deceased biologic father, and administration of own HIV medication were the main predictors of the disclosure status of a child. Age remains the strongest predictor of disclosure across all studies. Caregivers consider children less than 5 years of age as too young both emotionally and cognitively to their illness and the implications thereof [2, 14]. Caregivers are more likely to disclose the HIV status to children over the age of 12 [19]. Interestingly, the mean age at disclosure of our study population is consistent with the mean age at disclosure to HIV-infected children in the USA recently reported by the Pediatric AIDS Clinical Trial Group study (PACTG 219C) [3]. In the PACTG 219C study, there was a significant decline in age at disclosure over time [3]. The authors attributed the decline in age at disclosure to a decline in social stigma surrounding HIV and the improved long-term survival because of the introduction of HAART in 1996. Most resource limited countries either have access or are at the verge of accessing the advances in antiretroviral therapy. If the lesions learnt in resource rich countries are incorporated appropriately into programs in resource limited countries, there may be a rapid decline in the age at disclosure over time.

The present study was an exploratory study to determine prevalence and pattern of disclosure at the first Pediatric HIV/AIDS Care Program in Ghana. To the best of our knowledge, this is one of the first studies from West Africa. It has several strengths compared to previous studies from the sub-region; uses a quantitative instrument and a modest sample size. However, the study had some limitations. As a cross-sectional study, the associations observed may not be causal. Also, we are missing information on the circumstances leading to disclosure in individual situations and lack of data on adherence to treatment. Furthermore, the study did not explore the benefits of disclosure such as adherence and clinical improvement in HIV disease due to the low prevalence of disclosure. Further studies are needed to determine whether our findings are representative of the situation in Ghana and other sub-Saharan African countries.

In conclusion our study as other studies from the sub-region reiterates the need to shift the emphasis from whether or not to disclose to the HIV-infected child to providing culturally sensitive support to caregivers and age-appropriate information on the HIV status of the child [20]. Pediatric HIV programs should have a systematic and a staged-approach protocol in place to provide support and skill set to facilitate disclosure of HIV status to HIV-infected children by caregivers.

Acknowledgments

We are grateful to the Pediatric HIV/AIDS Care Program team at Korle-Bu Teaching Hospital, and Mr. Kakra Adjei and Jonas Kusah for their technical support. We thank the children and families for their participation. The Pediatric HIV/AIDS Care Program is supported by funding from the Korle-Bu Teaching Hospital, the Ministry of Health and the National AIDS Control Program through the Global Fund for AIDS, TB, and Malaria. SK was supported by Down’s International Travel Fellowship and Yale School of Medicine Student Research Fellowship and EP was supported by a grant from the National Institute of Allergy and Infectious Disease (KO8AI074404).

Copyright information

© Springer Science+Business Media, LLC 2010