Objective: To compare subjective parental perceptions of short-term diabetes control with objective glycated hemoglobin (HbA1c) measurements in a pediatric population.
Methods: Parents of 159 children and adolescents with type 1 diabetes anonymously completed a questionnaire concerning demographics, parental perception of the child's recent glycemic control, concern about nocturnal blood glucose levels, and factors related to severity of the child's diabetes. Nonparametric correlations were performed among all variables and the child's current HbA1c measurement. Good short-term glycemic control was defined as an HbA1c value of < or = 8.5%.
Results: Five factors were found to correlate significantly with good glycemic control, and these variables were subjected to a logistic regression analysis. With use of a cutoff value of 0.58, results explained 26% of the variance and had an overall 71% correct classification rate. Children with good glycemic control were younger, had a shorter duration of diabetes, lived in families with higher socioeconomic status, were less likely to experience hospitalizations due to their diabetes, and had parents who "always worried" about nocturnal blood glucose variations. Parental perception and HbA1c values were not significantly related (P>0.19); 60% of the parents correctly perceived the status of diabetes in their children, and 40% misperceived the level of glycemic control. Further analysis suggested that accuracy of parental perception was highly dependent on the child's level of glycemic control. Parents whose children had lower HbA1c values (< or = 7.5%) correctly perceived the status of blood glucose control in their children, whereas parents whose children had poorer glycemic control (HbA1c > or = 9.5%) had inaccurate perceptions.
Conclusion: Subjective parental perceptions of diabetes control in their children may not correspond to objective assessments. Our findings suggest a need for more educational feedback concerning HbA1c levels and recommended goals for glycemic control.