Queensland University of Technology, Institute of Health and Biomedical Innovation, School of Public Health and Social Work, Brisbane, Australia.
Queensland University of Technology, Institute of Health and Biomedical Innovation, School of Public Health and Social Work, Brisbane, Australia.
Am J Prev Med. 2017 Mar;52(3):391-402. doi: 10.1016/j.amepre.2016.10.042. Epub 2017 Jan 7.
Existing evidence shows that text message interventions can produce short-term health behavior change. However, understanding is limited regarding intervention characteristics moderating this effect or the long-term effectiveness of text message interventions on behavior change after contact stops.
MEDLINE, PubMed Central, ERIC, PsycINFO, and Web of Science were searched for articles published between April 2008 and December 2014 that evaluated an intervention targeting preventive health behaviors, delivered primarily by text message.
Intervention development and design characteristics and research outcomes were evaluated for 51 studies. Thirty-five studies were included in a meta-analysis (conducted in 2015) examining overall effect size and moderators of effect size. The overall pooled effect of interventions was d=0.24 (95% CI=0.16, 0.32, p<0.001) using outcome data collected most proximal to intervention cessation. Seven studies collected data following a no-intervention maintenance period and showed a small but significant pooled maintenance effect (d=0.17, 95% CI=0.03, 0.31, p=0.017, k=7). Few variables significantly moderated intervention efficacy. Interventions that did not use a theoretic basis, used supplementary intervention components, and had a duration of 6-12 months were most effective. The specific behavior being targeted was not associated with differences in efficacy nor was tailoring, targeting, or personalization of text message content.
Text message interventions are capable of producing positive change in preventive health behaviors. Preliminary evidence indicates that these effects can be maintained after the intervention stops. The moderator analysis findings are at odds with previous research, suggesting a need to examine moderators at the behavior-specific level.
现有证据表明，短信干预可以在短期内改变健康行为。但是，对于调节这种效果的干预特征或在接触停止后短信干预对行为改变的长期效果的理解有限。
从 2008 年 4 月至 2014 年 12 月，我们在 MEDLINE、PubMed Central、ERIC、PsycINFO 和 Web of Science 上搜索了评估主要通过短信发送的预防健康行为干预措施的文章。
对 51 项研究的干预措施的开发和设计特征以及研究结果进行了评估。35 项研究被纳入 2015 年进行的一项荟萃分析，以检验总体效应大小和效应大小的调节因素。使用最接近干预结束时收集的结果数据，干预措施的总体综合效应为 d=0.24（95%CI=0.16,0.32，p<0.001）。7 项研究在没有干预的维持期后收集数据，显示出较小但显著的综合维持效应（d=0.17，95%CI=0.03,0.31，p=0.017，k=7）。很少有变量显著调节干预效果。未使用理论基础、使用补充干预成分且持续时间为 6-12 个月的干预措施最有效。目标行为与疗效差异无关，也与短信内容的定制、针对性或个性化无关。
短信干预能够对预防健康行为产生积极的变化。初步证据表明，这些效果在干预停止后仍能维持。对调节因素的分析结果与之前的研究结果不一致，这表明需要在特定行为层面上检查调节因素。
