ORIGINAL COMMUNICATION A simplified method for assessing physical activity level values for a country or study population MTL Vasconcellos 1 * and LA Anjos 2 1 Departamento de Metodologia, Fundac ¸a˜o Instituto Brasileiro de Geografia e Estatı´stica (IBGE), Rio de Janeiro, Brasil; and 2 Departamento de Nutric ¸a˜o Social, Universidade Federal Fluminense, Nitero´i, Brasil, and Escola Nacional de Sau´de Pu´blica, Fundac ¸a˜o Oswaldo Cruz, Rio de Janeiro, Brasil Objective: To describe a method to estimate physical activity level (PAL) values based on household surveys in which representative information of the population on the distribution of sex, age, occupation and daily working hours is obtained, and to compare them to the recommended PAL (RPAL) values. Design and subjects: Data of 267446 individuals of a probabilistic sample from a Brazilian food consumption and family budget survey were used to estimate two different sets of estimated PAL (EPAL) values: (1) considering sex, age group and geographical area and (2) the above plus expenditure class, defined by the quartiles of per capita expenditure. Results: The EPAL values derived by the proposed method indicated that: (l) the RPAL values for 18+y-old women represent higher energy requirements (ERs) than necessary, which could make these women become obese; (2) the RPAL values for 18 to 59-y-old males can only be used to calculate the mean for the total population, because their use for the rural and urban populations separately will lead to severe physiological effectsFunder- and overnutrition, respectively; and (3) the RPAL values for 10 to 17 and 60+y-old males and 11 to 17-y-old females will lead to undernutrition by not incorporating provisions for the occupational activities. Conclusions: Sex- and age-specific RPAL values are not appropriate for the calculation of ERs of the Brazilian population. PAL values must be estimated and constantly recalculated as soon as the required information becomes available. It is important that these EPAL values be disseminated so that more adequate ERs are calculated. European Journal of Clinical Nutrition (2003) 57, 1025–1033. doi:10.1038/sj.ejcn.1601638 Keywords: energy requirements; surveys; energy metabolism Introduction In 1985, the FAO/WHO/UNU Expert Consultation on energy and protein requirements changed the method of estimating energy requirements (ERs) from the average energy intake of healthy people to the daily energy expenditure (EE) of the population. This change was based on the observation that estimation based on energy intake yielded a circular argu- ment in which the intake was not the most appropriate way to maintain healthy weight and optimal levels of physical activity for health in the long run. Two approaches were suggested to achieve this goal (FAO/ WHO/UNU, 1985; James & Schofield, 1990): (1) a factorial method in which the daily activities and their duration are identified and converted to ERs with the use of multiples of basal metabolic rate (BMR) for each task; and (2) the use of one value of physical activity level (PALFdefined as EE divided by BMR in 24 h), which is based on the occupational PAL (OPAL) classified into three groups: light, moderate and heavy. James and Schofield (1990) summarised the PAL values recommended in the 1985 report (FAO/WHO/UNU, 1985) for the appropriate physical activity of adolescents and adults as population means, and specifically for adults considering the OPAL for a daily duration of 8h of labour. Contrary to several investigators who have suggested a revision of the recommended PAL (RPAL) values for adults, primarily because EE estimated by doubly labelled water *Correspondence: MTL Vasconcellos, Departamento de Metodologia, Fundac ¸a ˜o Instituto Brasileiro de Geografia e Estatı ´stica (IBGE), Av. Repu ´ blica do Chile, 500-101 andar, 20031-170, Rio de Janeiro, Brazil. E-mail: mtlv@ibge.gov.br Guarantor: MTL Vasconcellos. Contributors: MTLV was involved in the design and coordination of the Estudo Nacional da Despesa Familiar defined the PAL value estimation method and conducted all data analysis. Both authors determined the IEI associated to each occupation and contributed to the concept and writing of the manuscript. European Journal of Clinical Nutrition (2003) 57, 1025–1033 & 2003 Nature Publishing Group All rights reserved 0954-3007/03 $25.00 www.nature.com/ejcn