There were 37.3 million people or 17.42% people out of the total population in Indonesia who lived under the poverty line in year 2003 which constitute 20.23% of people in rural and 13.57% of the people in urban area (BPS, 2003). Poverty has a negative impact to the health and nutrient status of the population who lived under the poverty line. Poor people are more exposed to environmental risk (poor sanitation, unhealthy food, violence and natural disaster) and have less access to quality health care (WHO, 2000). Various approaches have been conducted to determine the poverty line. Badan Pusat Statistics/Central Bureau for Statistics (BPS/CBS) is one of the institutions that have developed the poverty line, which has been used nationally in Indonesia. The poverty line was defined as the expenditure value of the minimum food, so called food poverty line, and non-food commodities per capita per month (BPS, 1998). The food poverty line was derived by calculating the rupiah value of minimum standard for food adequately required by an individual, which was set at 2,100 calories per person per day. The rupiah value of 2,100 calories was derived from 52 selected food commodities mostly consumed by the reference population from the Susenas/National Health Survey data. Reference population is a group of people living just above the poverty line, and hence, their pattern of consumption can be used as a standard of minimum consumption. The selected food commodities were taking into account the difference among provinces but it did not distinguished between rural and urban areas. Therefore, the difference in values of expenditure spent for purchasing these food commodities refers to the differences in volume and price of each selected food item in both areas. However, some aspects were not considered in the calculation of this food poverty line. First, the estimation of the calorie content from the selected food commodities was not based on the quantities of food actually consumed by household member. It was only concerned on the amount of food used by the household. However, food used is not equivalent to the food consumed. Therefore, the calories content derived from the food used would overestimate the calories intake derived from the food actually consumed by the household member (Hasan, 1997). Secondly, the food poverty line developed by BPS was not measured independently. It was based on the consumption of the reference population. The consumption of the reference population was not necessarily adequate in terms of quantity and quality of the food because the reference population was determined by their non-food expenditure. It did not guarantee that people with higher non-food expenditure would consume adequate food. Third, nutrient or food consumption pattern of the population was correlated by the range of foods, which was locally produced or imported, available in an area and in a given season (Gopalan, 1984 in Hatma 2001).
