Plasma concentrations are frequently used for therapeutic drug monitoring of antiretroviral drugs. Dried blood spot sampling offers a patient-friendly and easy alternative to plasma sampling. However, dried blood spot concentrations are not necessarily equal to plasma concentrations and therefore the objective of this work was to establish the relationship between nevirapine and efavirenz dried blood spot and plasma concentrations to facilitate clinical implementation of dried blood spot sampling. Paired dried blood spot and plasma samples were obtained from 40 HIV-infected patients on nevirapine and 40 on efavirenz treatment. All samples were analysed using validated HPLC-tandem mass spectrometry methods for the two matrices. Theoretical plasma concentrations were calculated from dried blood spot concentrations using the formula [dried blood spot concentration/(1 - haematocrit)] × fraction bound to plasma proteins = plasma concentration. Linear regression and Bland-Altman analysis were used to compare the two methods. Dried blood spot and plasma concentrations of nevirapine and efavirenz correlated well (r(2) = 0.867 and 0.972, respectively), although efavirenz dried blood spot concentrations were 39.8% (SD 7.1%) lower than plasma concentrations. Theoretical plasma concentrations (using patient-specific haematocrit) of nevirapine and efavirenz were similar to measured plasma concentrations, with a mean difference between the two methods of 0.29 mg/L (SD 1.35 mg/L) and 0.08 mg/L (SD 0.31 mg/L), respectively. Dried blood spot concentrations of nevirapine and efavirenz were equal to plasma concentrations after correction for haematocrit and compound-specific plasma protein binding and can therefore be used in clinical practice.