OBJECTIVES: Diabetes affects 3%-5% of total UK population and insulin is the largest drug-class category used to treat the disease. A greater understanding of the impact of different economic and epidemiological variables on total insulin sales will help the healthcare system and pharmaceutical industry be more responsive to demand and cost. METHODS: Generalized least squares regression with period random effects was used on a pooled yearly data set (2001 to 2010) of variables. The dependent variable was total yearly sales for insulin. The explanatory variables included - size of population; incidence and prevalence of diabetes; estimated total prescription (Rx) for insulin; and employee compensation per capita. The analysis used yearly pooled, cross-sectional data from IDF and IHS Forecasting Database at different time points to account for the variation in different variables. The total population was obtained from OECD. The prevalence and incidence rates were obtained from IDF for 2001, 2003, 2007 and 2010; average of previous year’s data-points were used for years in which no data was provided. The main independent variable was the total yearly prescription rate for insulin, calculated from data derived from intrinsic Patient Flow Model. RESULTS: A direct correlation was found between estimated total Rx for insulin, total population, and prevalence rates for diabetes. The results can be summarized as: For every 1% rise in total estimated Rx for Insulin and total population there is a 71% and 48% increase in insulin sales, respectively. CONCLUSIONS: Based on our model, total Rx plays a major role in determining the total sales for insulin. From a policy perspective, it will support UK government’s diabetes related initiatives focusing on effective cost management.