Implementation of isoniazid preventive therapy among hiv infected children in three health facilities in Nairobi county

Abstract

Background: Approximately 9 million new cases of tuberculosis (TB) occur in the world every year, 15% of these occur in children less than 15years. In Kenya, children comprise 11% of all tuberculosis cases. Kenya is position 15 of the 22 high tuberculosis burden countries. 1, 2 Human immunodeficiency virus (HIV) infection increases the risk of progression from latent tuberculosis to active disease by 20 to 37 fold.1 HIV infected children have a much higher risk of active disease. Isoniazid preventive therapy has been shown to reduce tuberculosis related morbidity by 72% and mortality by 54%.24 In 2013, only 21% of countries globally and 14 out of the 41 high TB/HIV burden countries reported provision of isoniazid preventive therapy (IPT) to people living with HIV.1 Following the 2011 Kenyan policy, data is needed to inform IPT implementation and scale up in our local context. Objectives: To determine the level of uptake of isoniazid preventive therapy among HIV infected children in care at Kenyatta National Hospital. To evaluate knowledge and practice of health workers at Kenyatta National Hospital, Mbagathi District Hospital and Langata Health Centre regarding isoniazid preventive therapy in HIV infected children. Study Design: This was a mixed method research combining a cross-sectional survey with structured, in-depth, key informant interviews. Methods The study was carried out in paediatric wards, paediatric outpatient clinicsand paediatric comprehensive HIV care clinic (CCC) at Kenyatta National Hospital and Mbagathi District Hospital together with the paediatric outpatient clinics at Langata Health Centre. We enrolled children aged 1 to 15 years with documented HIV infection receiving care in Kenyatta National Hospital and consenting health care workers (HCW) in the pediatric units at the three study facilities. We excluded children with active tuberculosis (TB) together with those found to have contraindications to isoniazid. We carried out consecutive recruitment of all HIV infected children aged 1 to 15 years receiving care in the comprehensive care clinic and those admitted in the wards. A questionnaire was administered to eligible child-caregiver pairs to assess knowledge and prior or current use of isoniazid for TB prevention. Each child was screened for TB using the WHO algorithm for TB screening followed by a complete physical examination and review of available relevant investigations. We approached all HCW in the paediatric units during the study period. A self-administered questionnaire was given to consenting HCW. We also conducted key informant interviews of HCW in the KNH CCC. Data collected was entered into preformed Access database and analyzed using SPSS version 17.0. Results: We enrolled 111 children with a median age of 8 (IQR 6.7- 9.6) years. Among the children in the study 58.6% were male and 106 (96%) were on ART. Baseline CD4was available for 104 children, 65% of these had a count of <500cells/µL. Majority (79%) had a current absolute count of>500cells/µL. We observed that IPT uptake was53.2% (95% CI 43.9% - 62.4%) among eligible children. The study demonstrates good completion (88%) of IPT once initiated. Caregiver education was found to be associated with better uptake of IPT. Additionally, children whose caregivers had a history of being on IPT had an increased likelihood of having received IPT (p<0.001, OR 27.50).Increase in baseline CD4 count of the child was associated with an increase in the probability of a child receiving IPT (p = 0.007). Of the 66 HCW interviewed 77% were working at KNH, 16% at Mbagathi DH and 6% at Langata HC. The majority (47%) were postgraduate doctors working in the inpatient department. The median duration in the paediatric unit was 15 (IQR 3 – 30) months. Relatively few health workers expressed concern that isoniazid (INH) was not effective enough (2%) or that the side effects were too dangerous (28.8%). Half of the health workers had not prescribed INH within the previous year. Moreover, 19.7% indicate a preference to wait to see whether a patient develops active TB, which can then be treated. Conclusion: This study demonstrated poor implementation of existing IPT guidelines and good completion (88%) of IPT once initiated. Half of the health workers had not prescribed INH within the previous year. There is a need to scale up and strengthen IPT services.