Abstract Background The decayed, missing, and filled teeth (DMFT/dmft) index recommended by the World Health Organization (WHO), which measures the prevalence of caries based on the presence of cavitated caries lesions, is the most used dental caries index in epidemiological studies. Early diagnosis of noncavitated carious lesions enables preventive measures, which has the potential to prevent dental caries‐related morbidity and reduce the financial burden associated with restorative or rehabilitative dental care. The International Caries Detection and Assessment System (ICDAS II) incorporates both the cavitated and noncavitated carious lesions with acceptable reliability. Aim To compare dental caries prevalence based on ICDAS II and WHO criteria. Design A cross‐sectional study was conducted among 362 children visiting People's Dental College and Hospital, Nayabazar, Kathmandu, Nepal to study dental caries prevalence based on the ICDAS II and WHO criteria. Results Among the study population, 290 (90.34%) and 169 (68.42%) children had dental caries in primary and permanent teeth according to the ICDAS II criteria, whereas according to WHO criteria, 267 (83.18%) and 107 (43.32%) had dental caries in primary and permanent teeth, respectively. The prevalence of dental caries was significantly higher ( p < .001) according to ICDAS II criteria than the prevalence based on WHO criteria in both dentitions. Conclusion This study showed a significant difference in dental caries prevalence between the ICDAS II and WHO methods of caries diagnosis. The presence of noncavitated carious lesions was alarming. To enable detection of early/noncavitated carious lesions, ICDAS II rather than WHO criteria of caries diagnosis may be a more a valuable tool.