Background: Effectiveness and toxicity of childhood cancer treatments have never been evaluated in Kenya since introduction of structured care in the early seventies. Objective:To evaluate effectiveness and toxicity of two treatment protocols for Non- Hodgkin’s lymphoma (NHL). Design: Historical cohort study using medical records. Setting: Kenyatta National Hospital, a tertiary care and medical teaching hospital. Subjects: Children = 15 years with diagnosis of non-Hodgkin’s Lymphoma. Main outcome measures: Primary outcomes were median survival, event free survival and toxicity. Results: Out of 101 records, only 26 (25.7%) met inclusion criteria. Baseline characteristics were similar in the two treatment arms. Median survival was 0.75 months (95%CI=0.54- 0.96) and 1.0 months (95% CI=0.29-1.71) for short and long arm groups. There was no difference in event free survival and haematological toxicity. Conclusion: No clear difference in effectiveness and toxicity between the intensiveshort and the less aggressive long course chemotherapy regimens was evident. Though lack of difference may be attributed to the small sample size, suboptimal supportive care for intensive treatment would increase risk of toxic deaths. As the short course protocol did not demonstrate obvious deterioration of median and event free survival, a strong case may be made for a randomised clinical trial within a context of improved supportive care.