An observational study on growth in HIV-infected children and adolescents on antiretroviral therapy in Europe, with special reference to darunavir (TMC114-EPPICC)

Study type

• Study topic:
  Disease /health condition

  Other
• Study topic, other:
  Disease/Epidemiology study
• Study type:
  Non-interventional study
• Scope of the study:
  Assessment of risk minimisation measure implementation or effectiveness
• Data collection methods:
  Secondary use of data

Non-interventional study

• Non-interventional study design:
  Cohort

Study drug and medical condition

• Medical condition to be studied:
  HIV infection

Population studied

• Short description of the study population:
  All HIV-infected children aged <18 years at initiation of cART with at least one height measurement at the start of cART and at least one measurement ≥1 year subsequently.
• Age groups:
  • Children (2 to < 12 years)
  • Adolescents (12 to < 18 years)
• Special population of interest:
  Immunocompromised
• Estimated number of subjects:
  175

Study design details

• Main study objective:
  Describe characteristics of HIV-infected children ever received a darunavir (DRV) -containing regimen, describe the change in height for age z-score (HAZ) pre- and post-DRV exposure. Compare the change in HAZ in those taking a DRV-containing regimen to those taking other non-DRV containing regimens.
• Data analysis plan:
  Standard descriptive statistics to summarizedemographic, clinical and drug utilization data for children takingDRV-containing and non-DRV-containing regimens within theEPPICC data set. Data on height will be transformed to height forage z-score (HAZ) based on the WHO Child Growth Standards andWHO Reference 2007. Comparability of groups initiating DRV-containing and non-DRV containing cART will be assessed using propensity scores. Providing sufficient overlap between the 2 groups, change in HAZ score will be compared using multilevel regression models to take account of the hierarchical structure of the data (repeated HAZ measurements for each child, and children within different cohorts). A fractional polynomial will be used to model change in HAZ over time, adjusting for baseline covariates related to demographic, HIV disease, treatment history and anthropometric characteristics.