| Objective: We conducted a meta-analysis of randomized control trials to evaluatethe effectiveness of non-pharmacological intervention on metabolic control for type2diabetes mellitus.Methods: We searched the PubMed, EMBASE for studies published from1999to March2012to identify randomized control trials (RCTs) about the effectiveness ofnon-pharmacological intervention on metabolic control for patients with type2diabetes.Inclusion criteria were:①Data integrity of literature published in English;②Onlystudies on patients with type2diabetes with no limits on race, sex and age;③Randomized control trials (RCTs) irrespective of their allocation concealment orblinding methods;④Studies which compared the effectiveness of non-pharmacologicalintervention versus simple treatment lasted at least for three months;⑤HbA1c, fastingblood glucose, lipid profile, blood pressure, BMI and other metabolic indexes,self-management skills, patient satisfaction, degree of knowledge were included. Trialsin which the interventions did not involve non-pharmacological treatment or the articleswere only published as abstracts, non-RCTs, duplicate publication were excluded. Tworeviewers reviewed the titles and abstracts of identified articles first, and then examinedthe full-text version of selected articles to assess relevance to the research topic. Theauthors were contacted by e-mail to obtain the missing data. The quality of includedstudies was evaluated by JADAD Scale. The main information included Author’s name,year of publication, study area, study population, interventions, study duration, thebaseline case and the disease status of the two groups of patients, the effect indicators,self-management skills, degree of knowledge, patient satisfaction. Data of mean andstandard deviation of HbA1c, fasting blood glucose (FBG), postprandial2h bloodglucose (PBG), cholesterol (TC), triglycerides (TG), low-density lipoprotein (LDL-C), systolic blood pressure (SBP), diastolic blood pressure (SBP), BMI were analyzed withRevMan5.0software. According to different modes of intervention, the articles wereclassified into four groups: nurse-led structured management group, telemedicineintervention group, peer or family support group, and self-management interventiongroup.Results: Thirty eight RCTs involving9849type2diabetes patients were retrieved.Meta-analysis suggested that there were significant difference between thenon-pharmacological intervention groups and the control groups for HbA1c[WMD=-0.39,95%CI:(-0.47,-0.31), P<0.00001], FBG [WMD=-0.84,95%CI:(-1.04,-0.64), P<0.00001], TC [WMD=-0.23,95%CI:(-0.40--0.06), P=0.008], TG[WMD=-0.63,95%CI:(-1.16,-0.11), P=0.02], LDL-C [WMD=-0.32,95%CI:(-0.61,-0.03), P=0.03], SBP [WMD=-3.00,95%CI:(-4.55,-1.44), P=0.0002], and DBP[WMD=-1.65,95%CI:(-2.93,-0.36), P=0.01]. But there was no significant differencebetween the non-pharmacological intervention group and the control group for BMI[WMD=-0.48,95%CI:(-1.00,0.05), P=0.07]. Nurse-led structured management[WMD=-0.41,95%CI:(-0.61,-0.21), P<0.0001], telemedicine intervention[WMD=-0.37,95%CI:(-0.52,-0.22), P<0.00001], and self-management intervention[WMD=-0.37,95%CI:(-0.48,-0.26), P<0.00001] were more effective than controlgroup for HbA1c. But there was no significant difference between the peer or familysupport group and the control group for HbA1c [WMD=-0.08,95%CI:(-0.43,0.27),P=0.66].Conclusion: Non-pharmacological interventions are effective for control ofHbA1c, lipid profile, blood pressure and other relevant metabolic indexes. Structuredand comprehensive diabetes management is more effective than simple interventions onglycaemic control. Diabetes management by specialist nurses are effective in controlHbA1c, lipid profile, blood pressures. Mobile technology and telemedicine aid in themanagement of diabetes. The peer or family members support does not help inglycaemic control. |
PDF Full Text Request