The Outpatient Nutrition Clinic of the Center for Research and Development of Nutrition (CRDN) has developed a six-month package with the main objective to improve nutrition status of severely malnourished children in its surrounding areas since 1981. However, there has never been a study to analyze its effectiveness. Therefore, there is a need to analyze its effectiveness and factors associated with the outcome. A six-month longitudinal study involving of384 patients with severe malnutrition was conducted from 1982-1997. Kaplan Meier survival analysis was used to analyse the probability of nutritional improvement from severe to moderate or mild malnutrition, while Cox regression model was used to analyse the relative improvement associated with several associated factors. The study revealed that overall probability to remain severe malnutrition (PRSM) for 6 months was 35 % with the median of 12 weeks, which mean that there was an improvement of nutritional status for 65 % cases with the median time of 12 weeks. Analysis of Cox regression model showed that only upper respiratory infection (URI), age of the child, and type of malnutrition were significantly associated with relative improvement (RI) of severe malnutrition. RI of mild URI was 1.88 times, moderate URI was 1.61 times compared to cases without URI after controlled by other factors in the model. RI of kwashiorkor, marasmic-kwashiorkor, and marasmus was 584.06 (the extrem value of RI was probably due to only 5 cases of kwashiorkor), 10.18, and 1.46 times than no-marasmus-kwashiorkor after being controlled by other factors in the model. RI of children aged 12-35 months was 1.43 times and 36-59 months was 0.68 times after being controlled by other factors in the model.