2008 年 66 巻 2 号 p. 57-67
We systematically reviewed randomized controlled trials on the transtheoretical model-based exercise guide (TTMEG) with over a six-month follow up period which had been published during 1982-2007. One of our research purposes was to summarize new evidence on the long-term effectiveness of TTMEG. The second purpose was to examine the differences in results by two kinds of outcome measures: stages of change (SOC) and physical activity (PA) level. Among 16 studies obtained from PubMed and PsycINFO searches, we further focused on studies with over a six-month interval between evaluation and the end of intervention. Four of the five studies with SOC outcome showed a significant effect over six months from the end of intervention, while two studies with PA outcome did not. Five studies showed both SOC and PA outcomes within the study, and the results were consistent between the outcome measures in four studies. In the remaining one study, only SOC was significant because of the greater variability inherent in the self-reported minutes spent in weekly PA. An evaluation using SOC defined the exercise, and it could evaluate the continuity of exercise over six months as well. SOC is therefore considered the preferable outcome measure of the long-term effects of TTMEG.