Abstract Background Effective interventions to prevent mother-to-child HIV transmission (PMTCT) exist and when properly applied reduce the risk of vertical HIV transmission. As part of optimizing PMTCT in the Dutch Caribbean we developed a set of valid and applicable indicators in order to assess the quality of care in HIV-infected (pregnant) women and their newborns. Methods A multidisciplinary expert panel of 19 experts reviewed and prioritized recommendations extracted from locally used international PMTCT guidelines according to a 3-step-modified-Delphi procedure. Subsequently, the feasibility, sample size, inter-observer reliability, sensitivity to change and case mixed stability of the potential indicators were tested for a data set of 153 HIV-infected women, 108 pregnancies of HIV-infected women and 79 newborns of HIV-infected women in Aruba, Curaçao and St Maarten from 2000 to 2010. Results The panel selected and prioritized 13 potential indicators. Applicability could not be tested for 4 indicators regarding HIV-screening in pregnant women because of lack of data. Four indicators performed satisfactorily for Curaçao ('monitoring CD4-cell count', 'monitoring HIV-RNA levels', 'intrapartum antiretroviral therapy and infant prophylaxis if antepartum antiretroviral therapy was not received', 'scheduled caesarean delivery') and 3 for St Maarten ('monitoring CD4-cell count', 'monitoring HIV-RNA levels', 'discuss and provide combined antiretroviral therapy to all HIV-infected pregnant women') whilst none for Aruba. Conclusions A systemic evidence-and consensus-based approach was used to develop quality indicators in 3 Dutch Caribbean settings. The varying results of the applicability testing accentuate the necessity of applicability testing even in, at first, comparable settings.