Costa Beth M, Fitzgerald Kristine J, Jones Kay M, Dunning Am Trisha
School of Nursing, Faculty of Health, Medicine, Nursing and Behavioural Sciences, Deakin University, Waterfront Campus, Geelong, 3220 Victoria, Australia.
BMC Fam Pract. 2009 Nov 17;10:72. doi: 10.1186/1471-2296-10-72.
Information technology (IT) is increasingly being used in general practice to manage health care including type 2 diabetes. However, there is conflicting evidence about whether IT improves diabetes outcomes. This review of the literature about IT-based diabetes management interventions explores whether methodological issues such as sample characteristics, outcome measures, and mechanisms causing change in the outcome measures could explain some of the inconsistent findings evident in IT-based diabetes management studies.
Databases were searched using terms related to IT and diabetes management. Articles eligible for review evaluated an IT-based diabetes management intervention in general practice and were published between 1999 and 2009 inclusive in English. Studies that did not include outcome measures were excluded.
Four hundred and twenty-five articles were identified, sixteen met the inclusion criteria: eleven GP focussed and five patient focused interventions were evaluated. Nine were RCTs, five non-randomised control trials, and two single-sample before and after designs. Important sample characteristics such as diabetes type, familiarity with IT, and baseline diabetes knowledge were not addressed in any of the studies reviewed. All studies used HbA1c as a primary outcome measure, and nine reported a significant improvement in mean HbA1c over the study period; only two studies reported the HbA1c assay method. Five studies measured diabetes medications and two measured psychological outcomes. Patient lifestyle variables were not included in any of the studies reviewed. IT was the intervention method considered to effect changes in the outcome measures. Only two studies mentioned alternative possible causal mechanisms.
Several limitations could affect the outcomes of IT-based diabetes management interventions to an unknown degree. These limitations make it difficult to attribute changes solely to such interventions.
信息技术（IT）在全科医疗中越来越多地被用于管理包括2型糖尿病在内的医疗保健。然而，关于IT是否能改善糖尿病治疗效果，证据存在矛盾。这篇关于基于IT的糖尿病管理干预措施的文献综述探讨了诸如样本特征、结局指标以及导致结局指标变化的机制等方法学问题，是否可以解释基于IT的糖尿病管理研究中一些明显不一致的研究结果。
使用与IT和糖尿病管理相关的术语搜索数据库。符合综述条件的文章评估了全科医疗中基于IT的糖尿病管理干预措施，且于1999年至2009年期间以英文发表。未包括结局指标的研究被排除。
共识别出425篇文章，16篇符合纳入标准：评估了11项以全科医生为重点和5项以患者为重点的干预措施。9项为随机对照试验，5项为非随机对照试验，2项为单样本前后设计。在所审查的任何研究中均未涉及糖尿病类型、对IT的熟悉程度和基线糖尿病知识等重要样本特征。所有研究均将糖化血红蛋白（HbA1c）作为主要结局指标，9项研究报告在研究期间平均HbA1c有显著改善；只有2项研究报告了HbA1c检测方法。5项研究测量了糖尿病药物，2项研究测量了心理结局。在所审查的任何研究中均未纳入患者生活方式变量。IT是被认为会影响结局指标变化的干预方法。只有2项研究提到了其他可能的因果机制。
若干局限性可能在未知程度上影响基于IT的糖尿病管理干预措施的效果。这些局限性使得难以将变化完全归因于此类干预措施。
