Project a large-scale, patient longitudinal database to the U.S. insured population. The AHRQ Medical Expenditures Panel Survey (MEPS) was used as the basis for the adjustment methodology. MEPS are a source of data representing the cost and use of health insurance coverage, and are comprised of several large scale surveys of families, individuals, employers, and health care providers. First, we subset the data source to the study population, then used multivariate logistic regression to construct demographics and case-mix based weights that were applied to make the data similar to the national sample. The weight is derived using inverse of probability of existing in the database. To validate the weights, we randomly divided MEPS data into two parts; training set, and validating set. We used the training set to estimate the weights, then validated weights comparing standardized differences in terms of demographics and health status between the weighted and validating data sets. The following variables were used in the logistic regression: age group, gender, race, location, income levels and health status (Charlson Comorbidity Index and Chronic Conditions). i3 data were more likely to be male, older, chronic, and white (p=0.0000). Adjusted weight values for the Commercial group ranged from 1 to 51 with median 1.63, Medicaid 1 to 104 with median 1.03, and Medicare 1 to 61 with median 1.07. After applying adjusted weights, standardized differences in all confounders were less than 105. National projection of a large-scale, patient longitudinal database requires adjustment from not only demographic factors but also case-mix differences related to health status. The created weights successfully balanced the population in terms of co-morbid conditions and chronic conditions as well as demographic factors.