Background: Sub Saharan Africa hosts 90% of the worldís Human Immunodeficiency Virus infected children, and in 2007 an estimated 150,000 of these were Kenyan (UNAIDS, 2007). Baseline neuropsychological performance is a strong predictor of their diseaseís progression (Van Rie et al, 2006). There is no data on the baseline neurodevelopment (ND) status of Human Immunodeficiency Virus infected children in Kenya. Interventions to minimise morbidity and improve quality of life are routinely not instituted in Kenya. Aim: The aim was to determine the prevalence, spectrum and severity of ND delay among a cohort of HIV infected Kenyan eligible for HAART initiation but had not received HAART previously children at Kenyatta National Hospital Comprehensive Care Center. Methods: This was a prospective observational cohort study carried out at the Kenyatta National Hospital, comprehensive care center, in Nairobi, Kenya. Bayley Scales of Infant Development third edition (BSID III) was carried out on 36 highly active anti-retroviral therapy (HAART) naÔve HIV infected children aged between two weeks and 36 months, about to start HAART. These had met all eligibility criteria for HAART according to the KNH CCC protocols and were due to start the therapy. Study subjects were scheduled to have monthly bailey scales assessments to a maximum of at least six months post HAART. This was to enhance follow ñ up as only the initial and sixth month assessment was crucial for data analysis. The childrenvi underwent laboratory tests and other clinical services in accordance to the hospitalís standard of care for HIV children. Results: Thirty two - out of thirty - six children with HIV had neurodevelopment delay in at least one of the five constructs assessed by the BSID III at baseline. In descending order of frequency ND was most common in motor, language, cognitive and adaptive BSID III constructs respectively. The child aged 18 months and older demonstrated more frequent ND delay in all constructs measured except motor while secondary and above maternal education level was associated with worse cognitive performance of this cohort. Cognitive, language and social emotional scores improved in the 12 children who completed follow up for at least six months of HAART. The clinical significance of this data is the fact that even after 6 months of HAART patients still fell in the ND delayed category. This shows HAART alone is not sufficient to address ND problems. This data is of clinical significance but further research needs to be done to validate its statistical significance. HAART had no effect in improving motor delay in these twelve children. Conclusion: Neurodevelopment delay in this cohort of Kenyan children with HIV is prevalent; 83.3% motor delay, 66.6% language delay, 61.2% cognitive delay, 60.6% adaptive delay and 36.4% socio-emotional delay. All facets of ND delay as measured by the BSID III are affected. The magnitude of delay falls in the range that would benefit from early interventional rehabilitation. Data of the effect of highly active antiretroviral provision in this cohort is minimal but it shows that these drugs alone are not enough to manage the delay. Further research needs to be done to validate this point