Associations between Disordered Eating Behaviors and HbA1c in Young People with Type 1 Diabetes: A Systematic Review and Meta-analysis
journal contribution
posted on 2025-01-21, 04:58authored byKP Marks, J Aalders, S Liu, M Broadley, M Thastum, MB Jensen, EH Ibfelt, NH Birkebaek, F Pouwer
Background:
In type 1 diabetes, disordered eating behaviors (DEB) can adversely impact
HbA1c. Diabetes-adapted DEB questionnaires assess intentional insulin omission, whereas generic
questionnaires do not. Given the number of studies describing DEB-HbA1c associations published
over the past decade, an updated systematic review is warranted.
Objective:
The study aimed to examine the associations between DEBs assessed by generic and diabetes-
adapted questionnaires (and subscales) and HbA1c among young people (<29 years) with
type 1 diabetes.
Methods:
A systematic search was conducted in PubMed, Embase, PsycInfo, and CINAHL databases.
Observational studies examining associations between DEB as assessed by questionnaires
and HbA1c were included. Publication information, DEB and HbA1c characteristics, and DEBHbA1c
associations were extracted. Hedges’ g was calculated for mean HbA1c differences between
groups with and without DEB.
Results:
The systematic search yielded 733 reports, of which 39 reports representing 35 unique
studies met the inclusion criteria. Nineteen studies assessing DEB by diabetes-adapted questionnaires
(n=5,795) and seven using generic questionnaires (n=2,162) provided data for meta-analysis.
For diabetes-adapted questionnaires, DEB was associated with higher HbA1c (g=0.62 CI=0.52;
0.73) with a similar effect size when restricted to validated questionnaires (g=0.61; CI=0.50;
0.73). DEB was not associated with HbA1c for generic questionnaires (g=0.19; CI=-0.17; 0.55), but
significantly associated with higher HbA1c for validated generic questionnaires (g=0.32; 95%
CI=0.16-0.48). Participant and HbA1c collection characteristics were often inadequately described.
Conclusion:
Diabetes-adapted DEB questionnaires should be used in youth with type 1 diabetes
because they capture intentional insulin omission and are more strongly associated with HbA1c
than generic DEB questionnaires.