Blending Quality Improvement and Research Methods for Implementation Science, Part II: Analysis of the Quality of Implementation

This column is the second in a 3-part series on blending methods for quality improvement (QI) with traditional methods of experimental research for the purpose of producing a scientifically rigorous evaluation of implementation. In Part I, “Design and Data Collection,”¹  we described the first 3 steps for blending QI and research methods: step 1—begin with a focused topic; step 2—select qualitative and quantitative metrics; and step 3—identify time intervals for evaluation. The following information builds on Part I by describing the next 3 steps in the process for blending the observational data from QI processes with the experimental design of a randomized controlled intervention: step 4—choose the unit of randomization; step 5—evaluate the quality of implementation over time; and step 6—evaluate differences in the quality of implementation across sites. These 3 steps describe how to address the QI component of the blended approach. These steps assist in identifying which...