To determine the frequency of sports physical activity in the urban population of Peru and to identify the sociodemographic, economic, and environmental factors associated with that activity. This study utilized information collected by the country's National Household Survey (Encuesta Nacional de Hogares) in the second quarter of 1997. That Survey is overseen by Peru's National Institute of Statistics and Informatics (Instituto Nacional de Estadística e Informática). The Survey was based on a probabilistic, multistage sample that was stratified for all the urban areas of the country, which was divided into eight geographic regions: metropolitan Lima, northern coast, central coast, southern coast, northern mountains, central mountains, southern mountains, and jungle. In total, 14 913 homes were visited and 45 319 people at least 15 years of age were interviewed. The frequency of engaging in sports physical activity was classified as daily, every other day, weekly, or occasional. "Regular sports activity" (RSA) was defined as engaging in sports either every day or every other day. The preferences for and obstacles to sports practice were also examined. A descriptive analysis of the levels of RSA was carried out for gender, using the chi-square test. The factors associated with RSA were analyzed through conditional multiple logistic regression and analysis of residuals, multicollinearity, and interactions. The level of significance was set at P < 0.05. Practicing sports at least once a week was more common among men (44.5%) than among women (32.4%), and the same was true for RSA (12.8% versus 10.5%). The age group with the highest level of RSA was 50-55 years for men (20%), and 40-45 years for women (18%). RSA was most common in three geographic regions: jungle (15.3%), central mountains (12.8%), and central coast (12.1%). RSA was least common in two regions: southern mountains (9.7%) and metropolitan Lima (10.6%). The income bracket was not associated with RSA. However, other variables associated indirectly with the socioeconomic level - such as having more formal education, being employed, and having access to the Internet or cable television - and consumption of sports information were significantly and directly associated with RSA. The most frequent barriers to practicing sports were the lack of time, the lack of nearby sports infrastructure (playing fields or courts, etc.), and people's lack of interest. RSA on the part of the members of a household was significantly associated with RSA performed by the head of the household (male or female). RSA is limited in the urban areas of Peru. This is most true for persons who are less than 30 years old, for women, and for residents of the Lima metropolitan area. This low RSA level is a challenge for public health, and it confirms the need for promoting active lifestyles. More study is needed on the observed positive influence when the head of the household performs RSA and on the fact that RSA is more common in urban areas outside metropolitan Lima. These two findings should also be taken into consideration in designing specific interventions.