For now, China gives great impetus to the strategy of fertilizer zero growth. The actual application effects of many fertilizer reduction with efficiency increase methods have attracted more and more attentions. In this study, the nitrogen fertilizer reduction with efficiency increase methods in double-rice system in red soil region of South China were explored through field experiments carried out in a typical reddish paddy soil in Jiangxi Province. Six treatments were designed, including none N fertilizer control (CK), conventional N application rate(C1), N application with more times with 30 kg/hm² N reduction(C2), 20% slow-release urea application at one time with 30 kg/hm² N reduction (H1), 50% slow-release urea application at one time (H2) with 30 kg/hm² N reduction and 80% slow-release urea application at one time with 30 kg/hm² N reduction (H3). The results showed that C2, H1, H2, H3 with 30 kg/hm² nitrogen reduction had no significant difference in rice yields, effective panicle numbers, thousand-grain mass, nutrient contents in grains and straws, soil total and available nutrients compared with C1 (P>0.05). The total N absorption and N harvest index of C1, C2, H1, H2 and H3 also showed no significant differences (P>0.05). However, C2, H1, H2 and H3 displayed significant higher nitrogen partial factor productivity and nitrogen recovery efficiency than C1 (P<0.05). Compared with C1, the nitrogen partial factor productivity of C2, H1, H2 and H3 averagely increased by 19.0%, 17.8%, 19.9% and 24.5% and the nitrogen recovery efficiency of C2, H1, H2 and H3 averagely increased by 61.7%, 44.9%, 57.3% and 72.3% in three years, respectively. While there were no significant differences in nitrogen partial factor productivity and nitrogen recovery efficiency among H1, H2, H3 and C2 (P>0.05). According to this study, applying nitrogen fertilizer for more times and applying a proportion of slow-release urea at one time increase the nitrogen use efficiency in the same degree and are both effective nitrogen fertilizer reduction methods. Considering the high labor cost and the expensive slow-release urea, application 20% slow-release urea at one time is more operable nitrogen fertilizer reduction method. In the circumstances of 30 kg/hm2 reduction, ordinary urea applied with 20% slow-release urea can meet the needs of rice growth, and more percentage of slow-release urea (50% and 80%) cannot further increase rice yield and nitrogen use efficiency.