At present, there is a paucity of research studies that comprehensively investigate the factors causing vulvovaginitis in young females. The aim of this work was to propose an algorithm for predicting the risk of vulvovaginitis in young girls and adolescents. The study recruited 252 healthy girls, who were stratified into a few groups depending on their sexual development on the Tanner scale. The composition of vaginal microbiota was determined in all the participants using real-time polymerase chain reaction (PCR); distribution of allele and genotype frequencies was assessed for the polymorphic variants of genes coding for pro- and anti-inflammatory cytokines. Based on the obtained data, we created a functional model for predicting the risk of vulvovaginitis in girls at different stages of sexual development. Favorable risk factors for Tanner I girls included predominance of obligate anaerobes in vaginal microbiota and the polymorphic IL10 variant (С-819Т) homozygous for TT. The sensitivity of the model was 80%, its specificity was 78%. Favorable risk factors for prepubertal and pubertal girls included predominance of aerobes in the composition of vaginal microbiota and the presence of the TT allele in the polymorphic IL10 gene variant (С-3953Т). The sensitivity of the model was 58.3%, whereas specificity, 94.1%. This study provides the rationale conforming with the principles of evidence-based medicine for using prevention measures in the groups at risk for vulvovaginitis at young age. The proposed measures allowed us to reduce the relapse rate of bacterial vulvovaginitis threefold.