A majority of diabetes self-management programs have been shown to improve knowledge, attitude, practice, and health care outcomes. However, in the literature the underlying causal mechanisms for the improvement attributable to health education have not been explored, especially, how diabetes educational intervention may affect diabetes care outcomes. The purpose of the present study was to identify the causal mechanisms responsible for improved knowledge, attitude, practice and outcomes, so that educational interventions can be tailored efficiently and effectively to patients who are most likely to benefit from self-care management. The study used the knowledge, attitude, practice and outcome (KAP-O) framework. The specific purpose of the study was to examine the causes of variation in the outcomes of glycated hemoglobin (A1C), low-density lipoprotein cholesterol (LDLC), functional capacity (FC), and poor perceived health (PPH).An experimental study with a randomized control trial design involving 141 participants was conducted. The experimental group (N = 87) and control group (N = 52) were comparable in terms of demographics and major diagnoses. The experimental group received diabetes education. The control and experimental groups received usual customary care. Knowledge, attitude, practice, functional capacity and poor perceived health were measured before and after intervention using reliable and valid instruments. The study used a tailored attitudinal instrument.Glycated hemoglobin (A1C) and low-density lipoprotein cholesterol (LDLC) were measured before and after intervention. Multiple analytic strategies were applied to examine the experimental data.The four outcome variables of (A1C), (LDLC), (FC), and (PPH) did not constitute one common factor measurement model for outcome evaluation. Results of the Independent sample Revision