BACKGROUND An estimate of problem drug use prevalence is a substantial part of the monitoring of the epidemiological situation in drug use and its consequences and an important indicator for drug policy implementation. The capture-recapture method (CRM) is one of the most commonly used standard methods for this purpose worldwide. METHODS AND MATERIAL The CRM was used to estimate the numbers of problem drug users (PDU) and problem users of opiates/opioids (PUO) in the Czech Republic in 2006 and 2007. The following data sources were used: the General Health Insurance Company records of payments to out- and in-patient psychiatric care providers, records of admissions to psychiatric hospitals, replacement therapy register, and reports of newly diagnosed cases of viral hepatitis. Cases were defined as diagnoses F11, F15, and F19 according to the International Classification of Diseases, Tenth Revision (ICD-10) from the health insurance and admission records, the replacement therapy register covers PUO by definition, and injecting drug users were selected from reported cases of viral hepatitis. Log-linear analysis in Rcapture (R) was performed and the Akaike information criterion was used for model selection. RESULTS Altogether 12,882 and 13,505 individuals entered into analysis of PDUs and 5146 and 5409 individuals entered into analysis of PUO in 2006 and 2007, respectively. The estimates of PDUs were 23,900 (95% CI: 20,700-28,500) in 2006 and 31,000 (25,500-39,400) in 2007. The estimates of PUO were 6,864 (6,641-7,113) in 2006 and 7,096 (6,871-7,346) in 2007. The male/female ratio was 2.2/1. In PDU estimates, 83% were in the age range 15-34 and 17% were aged above 34; in PUO estimates, the respective rates were 80% and 20%. The prevalence rates of PDU in the age range 15-64 were 2.28 and 2.95 per 1,000 population in 2006 and 2007, respectively. The prevalence rates of PUO in the age range 15-64 were 0.65 and 0.67 per 1,000 in 2006 and 2007, respectively. The highest prevalence of both PDUs and PUO was observed in the Usti Region and Prague, followed by the Karlovy Vary, Plzeň, and Central Bohemia Regions. CONCLUSION PDU estimates in the Czech Republic obtained using the CRM are consistent with those derived by other methods. The major drawbacks are the quality of potential data sources and hassles in their acquisition and preparation. The interconnection of the available data sources also limits the validity and representativeness of estimates and therefore, another independent data source(s) should be sought in the future.