Aims: We set out to test the hypothesis that intentional hyperglycemia at work (henceforward IHW) mediates the link between diabetes distress and glycemic control in working adults with type 1 diabetes.

Methods: Clinical information on HbA1c, complications, diabetes duration and age was linked with self-report measures of psychosocial health and work life from people with type 1 diabetes of working age attending a specialist diabetes clinic in Denmark. Diabetes distress was assessed with a work-comprehensive measure encompassing the PAID-5 and two additional items assessing distress in reconciling diabetes and work life. IHW was assessed with one item. Glycemic control was assessed with HbA1c. Using structural equation modeling techniques we modelled a theory-based structural path model and tested the mediation hypothesis using bootstrap estimates and goodness-of-fit tests.

Results: The sample consisted of 1,126 working adults with type 1 diabetes. The model suggested that the effect of diabetes distress on glycemic control was fully mediated by IHW. Diabetes distress was associated with more frequent IHW (β = .23, p < .001) which in turn was associated with elevated HbA1c (β = .28, p < .001). The final model was able to account for 20% of the variance of IHW, and 11% of the variance in glycemic control. There was an excellent fit to the data according to established criteria (in parenthesis); X2/DF=3.3 (<5), GFI=.995 (=.95), CFI=.990 (=.95), RMSEA=.047 [.030-.065] (<.050).

Discussion: This study highlights the role of IHW as a mediator for the link between diabetes distress and glycemic control specific to working adults with type 1 diabetes. A sound understanding of the mechanisms linking diabetes distress to suboptimal glycemic control in working people with type 1 diabetes can inform intervention development and ultimately increase the chance of improving diabetes self-care and quality of life of working people with type 1 diabetes.


U.M. Hansen: None. K. Olesen: None. T.C. Skinner: Board Member; Self; Decision Support Analytics. I. Willaing: None.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at