This paper assesses the standard of living using health-relatedoutcomes instead of income-based measures. We use the Body Mass Index(BMI) as a categorical variable to take into account how far people are froma normal range associated to a desirable quality of life. This is a proxy of theStandard of Living Index for the Colombian case. An OrderedProbitModelis used to determine the probability of being within that normal range or not,with two different specifications: the normative one that takes into accountthe traditional categories in which the BMI is ranked,and the relative one, inwhich the distribution of the sample is incorporated. We found that educationand a constructed index of living conditions have a significant effect on thedependent variable and that there are non-linear effects. We also found thatpeople with walking difficulties and adults have less probability of having anormal BMI.