Constructing a social vulnerability index (SVI) for Medellín's disabled population during 2008 aimed at determining areas which were reducing opportunities for this population to use their tangible and intangible assets, thus impairing their quality of life.This descriptive cross-sectional study drew on a source of secondary information regarding people having some kind of limitation recorded in the Quality of Life Survey, 2008. Physical, human and social variables were grouped when constructing the SVI; the models were run in principal component analysis to determine their degree of vulnerability, defined by the number of negative factors identified (high category=4 or 5, medium=2 or 3 and low=1 or none). Such classification led to identifying non-causal relationships with demographic variables through Mann-Whitney, Chi-square and Kruskal-Wallis tests (5.0 % statistical significance level); multinomial logistic regression was used for calculating adjusted measures for epidemiological measurement, such as opportunity ratios and confidence intervals.A degree of medium vulnerability predominated in disabled people living in Medellín (60.3 %) followed by low vulnerability (28.7 %) and high vulnerability populations (11.0 %).The proposed ISV classified the city's communes according to high, medium or low vulnerability, supported by the use of statistical and spatial location techniques.