This study investigates the effects of three factors: (1) use of a mixture of two different grades of hydroxypropyl methylcellulose (HPMC), (2) apparent viscosity, and (3) tablet hardness on drug release profiles of extended-release matrix tablets. The lot-to-lot apparent viscosity difference of HPMC K15M on in vitro dissolution was also investigated. Four test formulations were made, each containing 10% of a very water-soluble active pharmaceutical ingredient (API), 32% HPMC K15M, or a mixture of HPMC K100LV and HPMC K100M, 56% diluents, and 2% lubricants. Each formulation was made at two hardness levels. A 2(3) full factorial design was used to study various combinations of the three factors using eight experiments conducted in a randomized order. Dissolution studies were performed in USP apparatus I. The values of t50% (time in which 50% drug is released) and tlag (lag time, the time taken by the matrix tablet edges to get hydrated and achieve a state of quasi-equilibrium before erosion and the advance of solvent front through the matrix occur) were calculated from each dissolution profile. The similarity factor (f2) was also calculated for each dissolution profile against the target dissolution profile. A simple Higuchi-type equation was used to analyze the drug release profiles. Statistical analysis using analysis of variance (ANOVA) and similarity factor (f2) values calculated from the data indicated no significant difference among the t50% values and dissolution profiles respectively for all formulations. Within the 3.3-6 kp hardness range investigated, dissolution rates were found to be independent of tablet hardness for all the formulations. Although significantly shorter lag times were observed for the tablets formulated with low- and high-viscosity HPMC mixtures in comparison to those containing a single grade of HPMC, this change had no significant impact on the overall dissolution profiles indicated by the similarity factor f2 values. From this study it can be concluded that lot-to-lot variability in apparent viscosity of HPMC should not be a concern in achieving similar dissolution profiles. Also, results indicated that within the viscosity range studied (12,000-19,500 cps) an HPMC mixture of two viscosity grades can be substituted for another HPMC grade if the apparent viscosity is comparable. Also, the drug release is diffusion-controlled and depends mostly on the viscosity of the gel layer formed.