The purpose of this study was to determine whether integrating encoding instruction with reading instruction provides stronger gains for students who struggle with reading than instruction that includes little or no encoding. An instructional design model was investigated to best fit the data of 175 first-grade readers at risk for reading disabilities. Using cross-classified hierarchical linear modeling, variance in students' posttest scores could adequately be explained by students' initial encoding and decoding abilities, classroom and intervention encoding instruction time, and the number of supplemental integrated encoding and decoding intervention lessons received. Results indicated that integrating encoding and decoding instruction in first-grade classrooms, as well as supplemental intervention programs, may be the missing link to decreasing and possibly preventing future reading failure for students previously at risk for reading disabilities. (Contains 4 notes, 1 figure, and 4 tables.)