University of California, San Francisco Logo

University of California, San Francisco | About UCSF | Search UCSF | UCSF Medical Center

The hidden harm of home-based care: Pulmonary tuberculosis symptoms among children providing home medical care to HIV/AIDS-affected adults in South Africa
Global Health Sciences Literature Digest
Published March 27, 2013
Journal Article

Cluver L, Orkin M, Moshabela M, Kuo C, Boyes M. The hidden harm of home-based care: Pulmonary tuberculosis symptoms among children providing home medical care to HIV/AIDS-affected adults in South Africa. AIDS Care. 2013;25(6):748-55.


To determine whether children caring for HIV-infected adults had higher odds of developing pulmonary tuberculosis (TB) symptoms.


Three South African provinces with high HIV prevalence and high TB burden.

Study Design

Cross-sectional community-based survey.


Children aged 10-17, living in homes where they cared for one or more HIV-infected adults.

Main Outcome Measures

Pulmonary TB symptoms.


The study was conducted in Mpumalanga, KwaZulu-Natal and Western Cape, South Africa's provinces with the highest TB burden. Within these provinces, six health districts with >30% antenatal HIV prevalence were selected. To obtain the study sample, investigators performed a stratified random sampling of census enumeration areas or designated tribal areas within the six health districts. They then sampled households consecutively by going door to door. Sites ranged across a spectrum of settings from deep rural through dense rural, farming rural, peri-urban, urban and urban-homeland settings. One child was selected from each eligible household that consented to participate. Children were interviewed at home, at community centers or at their schools. Experienced interviewers assisted children in completing 40- to 60-minute questionnaires in the language of their choice (including isiXhosa, isiZulu, siSwati, Sesotho and Xitsonga). Questionnaires included a symptom-based TB checklist and also included questions based on investigators' hypotheses of risk factors for TB infection. The eight-item symptom checklist was based on models developed by the World Health Organization (WHO)(1) and the South African National Department of Health.(2) TB symptoms were considered to be severe if there were four or more symptoms of fever, discolored sputum, fatigue, weight loss and night sweats, in addition to two or more symptoms of coughing blood, chest pains and cough for >3 weeks.(3) Children and their families reporting severe TB symptoms were supported in accessing health services.

All survey items had been pre-piloted by an advisory group comprised of children, then piloted independently in child focus groups, and then translated and back-translated into the respective local languages.

Investigators hypothesized that risk factors for TB in children caring for HIV-infected adults included household poverty, food insecurity and household overcrowding, adult illness status (HIV-associated illness, other chronic illness, healthy) and child orphan status (AIDS-orphaned, orphaned by other causes and non-orphaned). Other hypothesized risk factors included child provision of personal or medical care within households and exposure to body fluids while providing care. Investigators performed multivariate logistic regression analyses, controlling for co-variates of urban or rural setting, gender, age and whether the children's primary caregivers were their biological parents.


A total of 6,002 children participated in the survey. Their mean age was 13.5 years (standard deviation [SD]=2.8), with slightly more than half (55.6%) females. Half lived in rural areas. Thirty percent lived with a non-parental guardian, and 45% had lost parents to AIDS. One quarter lived with guardians with AIDS-related illness, and 14% lived with guardians with other serious illness (e.g. cancers or diabetes). While the number of children actually providing care for parents or guardians with AIDS illness is not presented separately from the number of those actually caring for parents or guardians with other serious illness, 762 (12.7%) children cared for an unwell person, and 19.6% had been exposed to body fluids while providing care.

Overall, 285 (4.7%) had symptoms of pulmonary TB. TB symptoms were independently predicted by primary parent or guardian's AIDS-related illness (odds ratio [OR] 1.63, 95% confidence interval [CI] 1.23 to 2.15, p<0.001), but not by other parent or guardian illnesses or by having a healthy parent or guardian. TB symptoms in children were independently predicted by AIDS-orphanhood (OR 1.44, 95% CI 1.04 to 2.00, p<0.029), but not by orphanhood by other causes, or non-orphanhood. Household crowding was also a significant factor in predicting TB symptoms (OR 1.35, 95% CI 1.06 to 1.72, p<0.017) as was child food insecurity (OR 1.52, 95% CI 1.15 to 2.02, p<0.003). Although children's provision of care to sick persons, per se, was not predictive of TB symptoms, children exposed to patient body fluids while providing care had more than three times the odds of having TB symptoms (OR 3.12, 95% CI 1.96 to 4.95, p<0.001).

There was a clear gradient of increasing odds for TB symptoms with higher numbers of concurrent risk factors. Only 1.4% of children with no risk factors had TB symptoms, while 18.1% with all risk factors (i.e. AIDS-orphaned, living in a crowded home with a guardian suffering AIDS-related illness, exposed to body fluids through care provision, and lacking sufficient food) had symptoms. Most (75%) children with TB symptoms had never been tested for TB, and very few (11%) reported ever received TB treatment. Numerators for this paragraph's analyses were not provided.

Children's HIV serostatus was not assessed, but the proportion of HIV-infected children was assumed to be very small due to the availability of antiretroviral prophylaxis through most of the children's lives, and low child survival rates in the pre-antiretroviral era.

Table: Predictors of child pulmonary tuberculosis symptoms (from the article)


The authors conclude that there is an increased risk for symptoms of pulmonary TB among children affected by familial AIDS, heightened by exposure to body fluids during care provision to family members with AIDS-related illness and made worse by socioeconomic factors of living in crowded homes with insufficient food. They say that it is important to recognize role of children as caregivers and suggest that it may be possible to reduce child health risks by providing families with information, protective equipment to reduce exposure to body fluids and additional home-based support.

Risk of Bias

The risk of bias in this study is moderate to high. As a cross-sectional survey, there was no intervention and no comparison group. The survey had a large sample size, however and was conducted in diverse but high-burden and low-resource settings. The survey methodology was strong, although TB symptoms were not confirmed by sputum or smear screening tests. A study in Zimbabwe found that provider-initiated symptom screening in HIV-uninfected adults and observing "any symptom" (cough, blood in the sputum, fever, night sweats or weight loss), was better for ruling out TB than for diagnosing it, with 71% sensitivity and 90.3% specificity.(4)

In Context

Pediatric TB is acquired predominantly through contact with TB-infected adults.(5, 6) Untreated pulmonary TB in childhood often leads to death or long-term disability. Most children in sub-Saharan Africa with TB are not tested for it and their cases are not reported.(7) Consistent with the findings of this study, other studies in southern Africa have identified several risk factors for TB in children, including crowded homes and food insecurity,(7) being orphaned (among HIV-infected children),(9) and exposure time, home ventilation, and the period of adult infectivity.(10)

Programmatic Implications

Children in settings with high TB and HIV burden should be offered symptom-based screening as well as preventive chemotherapy. Interventions to improve TB referral systems, increase uptake of TB testing and treatment and improve infection control in households are needed.


  1. World Health Organization. Practical approach to lung health (PAL): A primary health care strategy for the integrated management of respiratory conditions in people five years of age or over (2005). [accessed 19 March 2013]
  2. National Department of Health, South Africa. The National Infection Prevention and Control Policy for TB, MDRTB and XDRTB (2007). [accessed 19 March 2013]
  3. Shah S, Demissie M, Lambert L, et al. Intensified tuberculosis case finding among HIV-infected persons from a voluntary counseling and testing center in Addis Ababa, Ethiopia. J Acquir Immune Defic Syndr. 2009 Apr 15;50(5):537-45.
  4. Corbett EL, Zezai A, Cheung YB, Bandason T, Dauya E, Munyati SS, Butterworth AE, Rusikaniko S, Churchyard GJ, Mungofa S, Hayes RJ, Mason PR. Provider-initiated symptom screening for tuberculosis in Zimbabwe: diagnostic value and the effect of HIV status. Bull World Health Organ. 2010 Jan;88(1):13-21.
  5. Loeffler AM. Pediatric tuberculosis. Semin Respir Infect. 2003 Dec;18(4):272-91.
  6. Schneider H, Lehmann U. Lay health workers and HIV programmes: implications for health systems. AIDS Care. 2010;22 Suppl 1:60-7.
  7. Marais BJ, Obihara CC, Warren RM, Schaaf HS, Gie RP, Donald PR. The burden of childhood tuberculosis: a public health perspective. Int J Tuberc Lung Dis. 2005 Dec;9(12):1305-13.
  8. Beisel WR. Nutrition in pediatric HIV infection: setting the research agenda. Nutrition and immune function: overview. J Nutr. 1996 Oct;126(10 Suppl):2611S-2615S.
  9. Braitstein P, Nyandiko W, Vreeman R, Wools-Kaloustian K, Sang E, Musick B, Sidle J, Yiannoutsos C, Ayaya S, Carter EJ. The clinical burden of tuberculosis among human immunodeficiency virus-infected children in Western Kenya and the impact of combination antiretroviral treatment. Pediatr Infect Dis J. 2009 Jul;28(7):626-32.
  10. Wood R, Johnstone-Robertson S, Uys P, Hargrove J, Middelkoop K, Lawn SD, Bekker LG. Tuberculosis transmission to young children in a South African community: modeling household and community infection risks. Clin Infect Dis.2010 Aug 15;51(4):401-8. doi: 10.1086/655129.