OBJECTIVE : To validate the Brazilian version of the Short Assessment of Health Literacy in Portuguese-speaking Adults (SAHLPA), a 50-item test proposed as a particularly helpful instrument to assess health literacy in people with limited skills, in the Portuguese population. METHODS : We used the standard procedure for cultural adaptation and administered the instrument to 249 participants. We examined construct validity using groups with expectedly increasing levels of health literacy (laypersons from the general population, engineering researchers, health researchers, and physicians), and through association with age and educational attainment, dichotomizing scores at the median of the layperson's group. RESULTS : Exploratory factor analysis revealed the instrument was one-dimensional and justified reduction to 33 items. SAHLPA-33 displayed adequate reliability (Cronbach's α = 0.73). The frequency of limited health literacy was highest among laypersons and lowest among physicians (p <0.001; p for trend <0.001). The proportion of participants with limited health literacy decreased with increasing education attainment (age- and sex-adjusted p for trend <0.001). Limited health literacy also tended to decrease with age, although the association was non-significant (sex- and education-adjusted p for trend = 0.067). CONCLUSIONS : We culturally adapted a brief and simple instrument for health literacy assessment, and showed it was valid and fairly reliable. In Portuguese low-literate adults, SAHLPA-33 fills the gap in health literacy assessment instruments, and may be used to guide communication strategies with vulnerable patients and communities.