This case control study was aimed to identify the predictive risk factors for frequent relapse idiopathic nephrotic syndrome (INS) and conducted in Sir Salimullah Medical College & Mitford Hospital, Dhaka and at Renal and Dialysis Unit of Dhaka Shishu Hospital and Bangladesh Institute of Child Health (BICH), Dhaka, from January 2006 to December 2006. We examined retrospectively the clinical course of fifty cases of frequent relapse nephrotic syndrome (FRNS) as cases and fifty cases of infrequent relapse nephrotic syndrome (IRNS) as control who met the predefined enrollment criteria, followed for at least one year after initial onset of disease. After enrollment following parameters were studied as predictors of frequent relapse: i) Socio-demographic variables: age, sex, socio-economic condition, number of living room ii) Disease related variables i.e. age of onset, duration of illness, frequency of relapse within the 1st year, regimen of initial steroid therapy, total cumulative dose of steroid for remission, day of remission after starting steroid, association with atopy and infection, concomitant upper respiratory illness iii) Biochemical and pathological variables (at the time of initial attack) i.e. Serum albumin, serum cholesterol, blood urea, 24 hours urinary protein, serum creatinine, complete blood count, urine RBC, urine pus cell, urine culture. The test statistics used to analyses the data were descriptive statistics, Chi-square probability test, Student's t-test and Binary logistic regression analysis for Odds ratio. Both univariate and multivariate logistic regression analysis revealed that age of onset (P<0.001, OR=0.9, 95% CI=0.85-0.95), poor socioeconomic status (P<0.034, OR=0.5.8, 95% CI=1.14-29.5) and low serum albumin level at the time of initial presentation (P<0.022, OR=0.8, 95% CI=0.65-0.97) were independent predictors of frequent relapse nephrotic syndrome. In conclusion, we demonstrated that age at onset, poor socioeconomic condition and low serum albumin level at the time of initial attack is independent risk factors for frequent relapse idiopathic nephrotic children. This information can be used as a predictor of FRNS.