Total Cholesterol Level among Filipino Adults using Small Area Estimation

Lyndon Morales

Health-related data are being collected from a large number of surveys that give health information at the national level. These are aggregate data from census and usually lack detail. The demand for reliable small area statistics is growing, specifically for government intervention and policy-making. Direct survey estimates for small areas are likely to yield unacceptable large standard errors due to smallness of sample sizes. Alternative methods for describing a smaller population have been evaluated by relating areas to each other to arrive at more accurate estimates.

Projects that need evaluation for implementation and monitoring make use of statistics in the specific provinces or municipalities where more government support should be given. These statistics coming from Nationwide surveys like Family Income Expenditure Survey to be used as a measure for the local geographical divisions usually lead to estimates with large standard errors, which may have been caused by the number of samples used per province that are not be enough to provide specific accuracy at a given domain. In this reason, the issue on decrease in error of estimates may be addressed without the increase in the cost of conducting a nationwide survey via the application of small area estimation techniques.

According to the World Health Organization, the magnitude of cardiovascular diseases (CVD’s) is the number 1 cause of death, wherein 31% of global deaths is due to CVD’s (WHO, 2017). In the Philippines, diseases of the heart and the cerebrovascular diseases (stroke) remain the top two leading causes of mortality in 2013. In addition, the overall mortality of Filipinos shows an increasing trend in the mortality rate due to cardiac diseases in 2013 (PSA, 2016).

High cholesterol level, together with other factors, is one of the important risk factor of CVD. Behavioral risk factors can be measured through the determination of the lipid profile and presence of other risk factors, which indicates high risks of having detrimental cardiovascular events (WHO Factsheet, 2017).

A study should be done to compare Individual Level Synthetic Regression and Relational Model in generating small area estimates and estimate the prevalence of high total cholesterol level in Pampanga, Region III. Data can be utilized from the 2013 National Nutrition Survey (NNS) conducted by the Food and Nutrition Research Institute under the Department of Science and Technology. The survey participants in the 8th NNS represent the 17 regions, 79 provinces, 45,047 households and 172,323 individuals. The 8th NNS was conducted from June 2013 to April 2014. The said data is from the Clinical Health Survey Component which consists of the study population with adult Filipinos aged 20 years and above.

--oOo-