Sharleen L O'Reilly, James A Dunbar, Vincent Versace, Edward Janus, James D Best, Rob Carter, Jeremy J N Oats, Timothy Skinner, Michael Ackland, Paddy A Phillips, Peter R Ebeling, John Reynolds, Sophy T F Shih, Virginia Hagger, Michael Coates, Carol Wildey, MAGDA Study Group, Wareham, Nicholas J, and Lee Kong Chian School of Medicine (LKCMedicine)
Background Gestational diabetes mellitus (GDM) is an increasingly prevalent risk factor for type 2 diabetes. We evaluated the effectiveness of a group-based lifestyle modification program in mothers with prior GDM within their first postnatal year. Methods and Findings In this study, 573 women were randomised to either the intervention (n = 284) or usual care (n = 289). At baseline, 10% had impaired glucose tolerance and 2% impaired fasting glucose. The diabetes prevention intervention comprised one individual session, five group sessions, and two telephone sessions. Primary outcomes were changes in diabetes risk factors (weight, waist circumference, and fasting blood glucose), and secondary outcomes included achievement of lifestyle modification goals and changes in depression score and cardiovascular disease risk factors. The mean changes (intention-to-treat [ITT] analysis) over 12 mo were as follows: −0.23 kg body weight in intervention group (95% CI −0.89, 0.43) compared with +0.72 kg in usual care group (95% CI 0.09, 1.35) (change difference −0.95 kg, 95% CI −1.87, −0.04; group by treatment interaction p = 0.04); −2.24 cm waist measurement in intervention group (95% CI −3.01, −1.42) compared with −1.74 cm in usual care group (95% CI −2.52, −0.96) (change difference −0.50 cm, 95% CI −1.63, 0.63; group by treatment interaction p = 0.389); and +0.18 mmol/l fasting blood glucose in intervention group (95% CI 0.11, 0.24) compared with +0.22 mmol/l in usual care group (95% CI 0.16, 0.29) (change difference −0.05 mmol/l, 95% CI −0.14, 0.05; group by treatment interaction p = 0.331). Only 10% of women attended all sessions, 53% attended one individual and at least one group session, and 34% attended no sessions. Loss to follow-up was 27% and 21% for the intervention and control groups, respectively, primarily due to subsequent pregnancies. Study limitations include low exposure to the full intervention and glucose metabolism profiles being near normal at baseline. Conclusions Although a 1-kg weight difference has the potential to be significant for reducing diabetes risk, the level of engagement during the first postnatal year was low. Further research is needed to improve engagement, including participant involvement in study design; it is potentially more effective to implement annual diabetes screening until women develop prediabetes before offering an intervention. Trial Registration Australian New Zealand Clinical Trials Registry ACTRN12610000338066, Sharleen O'Reilly and colleagues investigate the effectiveness of a diabetes prevention program for reducing weight, waist circumference and fasting glucose measurements for women who have had gestational diabetes., Author Summary Why Was This Study Done? Women who have had gestational diabetes are much more likely to develop type 2 diabetes. Although many diabetes prevention programs for people over the age of 50 exist, few are tailored to the needs of young mothers who have had gestational diabetes. On the assumption that offering prevention earlier is beneficial, researchers developed and tested a diabetes prevention program for women who had gestational diabetes; women participated in the program during their first year after giving birth. What Did the Researchers Do and Find? The researchers enrolled 573 women in a one-year study: 284 women were assigned to the diabetes prevention program (one individual session and five group sessions over a three-month period, followed by telephone calls at six and nine months), and 289 were assigned to the control group (usual postnatal care). After one year, the average changes for women in the diabetes prevention program were a 0.23-kg decrease in weight, a 2.24-cm decrease in waist circumference, and a 0.18-mmol/l increase in fasting blood glucose, while the average changes for women in the control group were a 0.72-kg increase in weight, a 1.74-cm decrease in waist circumference, and a 0.22-mmol/l increase in fasting blood glucose. The between-group difference in weight change was 0.95 kg. The number of women who attended the diabetes prevention program was lower than anticipated—10% attended all sessions, and 53% attended the individual session plus at least one group session; about a quarter of women did not complete the study, mainly due to becoming pregnant again. What Do These Findings Mean? These findings suggest that although a diabetes prevention program designed for women who have had gestational diabetes can prevent weight gain over 12 months, getting women to engage with the program was challenging, so it would not be sustainable in routine health services. The women who participated in the study had low diabetes risk profiles (only one in ten had impaired glucose tolerance), and most diabetes prevention guidelines would not categorise them as being at sufficiently high risk for participation in a diabetes prevention program. For diabetes prevention programs in women who have had gestational diabetes, further research is required on the process of engagement and lifestyle interventions at other time points, including participant involvement in the design of interventions. Australian clinical guidelines stipulate that women who have had gestational diabetes should be screened annually for diabetes. One option for management would be to wait until they develop prediabetes before offering a diabetes prevention program, which may prove more effective because their children will be older and women may be easier to engage in improving their health.