N W Cheung1. 1. Centre for Diabetes & Endocrinology Research, Westmead Hospital, Westmead, NSW 2145, Australia. wah@westgate.wh.usyd.edu.au
Abstract
AIM: Whether parity promotes the development of diabetes has been a subject of controversy. Earlier population studies have produced discrepant findings. The aim of the current study was to investigate this question by examining the relationship between parity and the age of diagnosis of diabetes. METHODS: A retrospective analysis from a diabetes database was conducted. Data for 2102 women with type 2 diabetes (T2DM), including date of birth, age at diagnosis, parity, body mass index (BMI), and family history were extracted. A multivariate linear regression model was applied to ascertain if there was an association between parity and the age of diagnosis of diabetes. RESULTS: With correction for the year of birth, there was a significant association between parity and the age of diagnosis of diabetes (beta = -0.25 +/- 0.08, P = 0.002, where beta is the regression coefficient). That is, for an increase in parity by one, the age of diagnosis decreased by 0.25 years. When further adjustments were made for the other variables in the data set, the effect on the overall cohort disappeared, but it persisted in subjects born before 1950 (beta = -0.17 +/- 0.1, P = 0.09), or with a parity >5 (beta = -0.60 +/- 0.25, P = 0.02). CONCLUSIONS: There is an association between parity and the age of diagnosis of diabetes, suggesting that pregnancy may promote the development of diabetes. However, the magnitude of the effect is small, and unlikely to be of great clinical significance in society today.
AIM: Whether parity promotes the development of diabetes has been a subject of controversy. Earlier population studies have produced discrepant findings. The aim of the current study was to investigate this question by examining the relationship between parity and the age of diagnosis of diabetes. METHODS: A retrospective analysis from a diabetes database was conducted. Data for 2102 women with type 2 diabetes (T2DM), including date of birth, age at diagnosis, parity, body mass index (BMI), and family history were extracted. A multivariate linear regression model was applied to ascertain if there was an association between parity and the age of diagnosis of diabetes. RESULTS: With correction for the year of birth, there was a significant association between parity and the age of diagnosis of diabetes (beta = -0.25 +/- 0.08, P = 0.002, where beta is the regression coefficient). That is, for an increase in parity by one, the age of diagnosis decreased by 0.25 years. When further adjustments were made for the other variables in the data set, the effect on the overall cohort disappeared, but it persisted in subjects born before 1950 (beta = -0.17 +/- 0.1, P = 0.09), or with a parity >5 (beta = -0.60 +/- 0.25, P = 0.02). CONCLUSIONS: There is an association between parity and the age of diagnosis of diabetes, suggesting that pregnancy may promote the development of diabetes. However, the magnitude of the effect is small, and unlikely to be of great clinical significance in society today.
Authors: Ditte Smed Iversen; Julie Støy; Ulla Kampmann; Thomas Schmidt Voss; Lene Ring Madsen; Niels Møller; Per Glud Ovesen Journal: BMJ Open Diabetes Res Care Date: 2016-08-25