Longitudinal trends in HbA1c patterns and association with outcomes: A systematic review

• Published in: Diabetes/metabolism research and reviews Vol. 34; no. 6; pp. e3015 - n/a
• Main Authors: Luo, Miyang, Tan, Kristin Hui Xian, Tan, Chuen Seng, Lim, Wei Yen, Tai, E‐Shyong, Venkataraman, Kavita
• Format: Journal Article
• Language: English
• Published: Bognor Regis Wiley Subscription Services, Inc 01.09.2018; John Wiley and Sons Inc
• Subjects: Cardiovascular diseases; Diabetes; Diabetes mellitus; Diabetes mellitus (insulin dependent); Diabetes mellitus (non-insulin dependent); diabetes‐related outcomes; Evidence-based medicine; glycaemic control; group‐based trajectory analysis; HbA1c; Insulin; longitudinal trends; Minority & ethnic groups; Patients; Population studies; Risk assessment; Risk factors; Systematic review; Trends
• ISSN: 1520-7552; 1520-7560; 1520-7560
• DOI: 10.1002/dmrr.3015

Background This study aimed to review studies that identified patterns of longitudinal HbA1c trends in patients with diabetes and to summarize factors and outcomes associated with distinct trajectory patterns. Methods PubMed and Web of Science were systematically searched for studies examining HbA1c trends among patients with diabetes from database inception through September 2017. Articles were included if they met the following inclusion criteria: (a) longitudinal study of subjects with diabetes only, (b) use of serial measurements of HbA1c, and (c) analysis of the trend of HbA1c using group‐based trajectory approaches. Results Twenty studies were included, 11 on type 1 diabetes and 9 on type 2 diabetes. These studies identified 2 to 6 HbA1c trajectory patterns. The most commonly identified patterns included stable HbA1c around 7.0% and at levels between 8.0% and 9.9%, which usually captured the HbA1c pattern among the majority of subjects in the study population. Unstable patterns identified included increasing HbA1c trend, decreasing HbA1c trend, and non‐linear patterns. These patterns were associated with differential risk of disease outcomes, over and beyond single‐point HbA1c measures. Age, gender, ethnicity, diabetes duration, disease management frequency, cardiovascular risk factors, insulin treatment, family environment, and psychosocial factors were the most frequently reported factors associated with membership of specific HbA1c pattern groups. Conclusion Common patterns of longitudinal HbA1c trends were identified despite heterogeneity among the studies. A better understanding of what underlies these different patterns may provide opportunities to tailor therapies and care for these patients to reduce adverse outcomes.