Meta-Analysis of Randomized Educational and Behavioral Interventions in Type 2 Diabetes

Tiffany L. Gary, Jeanine M. Genkinger, Eliseo Guallar, Mark Peyrot, Frederick L. Brancati

Research output: Contribution to journalArticle

Abstract

PURPOSE this meta-analysis was conducted to assess the effect of educational and behavioral interventions on body weight and glycemic control in type 2 diabetes. MEHODS Studies selected for analysis were published randomized controlled trials that evaluated educational and behavioral interventions (no drug interventions) in type 2 diabetes (sample size 210). These criteria were applied to searches of electronic databases and relevant bibliographies. Data were independently abstracted by 2 reviewers and adjudicated by consensus. RESULTS Of the 63 articles that met the inclusion criteria, 18 provided enough information for pooled estimates of glycohemoglobin (total Ghb, HbA1, or HbA1 C). These 18 studies yielded 2720 participants (sample sizes of 18 to 749). Interventions ranged from 1 to 19 months; follow-up ranged from 1 to 26 months. Glycohemoglobin was reduced by a mean of 0.43%. When results were stratified by quality score, glycohemoglobin was -0.50% and -0.38% for studies with high and low quality scores, respectively. When weighting studies by sample size, fasting blood glucose was reduced by 24 mg/dL and weight by 3 lbs. CONCLUSIONS Previous educational and behavioral interventions in type 2 diabetes have produced modest improvements in glycemic control. Future research should refine such interventions and improve methodology.

Original languageEnglish (US)
Pages (from-to)488-501
Number of pages14
JournalThe Diabetes Educator
Volume29
Issue number3
DOIs
StatePublished - May 2003

ASJC Scopus subject areas

  • Endocrinology, Diabetes and Metabolism
  • Health Professions (miscellaneous)

Fingerprint Dive into the research topics of 'Meta-Analysis of Randomized Educational and Behavioral Interventions in Type 2 Diabetes'. Together they form a unique fingerprint.

  • Cite this