PITCH: Factors associated with six-week postpartum glucose tolerance testing in women with a history of gestational diabetes

Problem

Gestational diabetes (GDM) is the most common pregnancy disorder and the single most important risk factor for type 2 diabetes (T2D). It is known that many women do not attend six-week postpartum screening for glucose abnormalities. Previous studies report inconsistent associations between completion of testing and factors including severity of GDM, parity, body mass index (BMI), and demographic characteristics. We aimed to describe the frequency of postpartum testing and patient- and general practice-related factors associated with attendance in a large sample of women with GDM.

Approach

We identified women diagnosed with GDM, according to the International Association of the Diabetes and Pregnancy Study Groups (IADPSG) criteria, between October 2014 and March 2017 from electronic medical records at Cambridge University Hospitals NHS Foundation Trust. We linked these records to practice characteristics and diabetes Quality and Outcomes Framework (QOF) indicators from Public Health England’s National General Practice Profiles using the most recent data available (usually 2016/17). Using STATA 15.1, we ran univariable and multivariable logistic regression analyses to determine which factors were associated with postpartum testing by an oral glucose tolerance test (OGTT) or HbA1c. We adjusted for clustering by practice and defined statistical significance as p<0.05. Odds ratios (OR) and 95% confidence intervals (CI) are reported.

Findings

Of 557 women with GDM, with mean age 33.7 (standard deviation 5.4) years, 415 (74.5%) had a postnatal test at 6.8 (2.1) weeks after delivery: 264 (63.6%) by OGTT, 44 (10.6%) by HbA1c and 107 (25.8%) had both tests. Two (0.5%) were diagnosed with T2D.

In univariable analyses, higher parity (OR 0.77, CI 0.66–0.91), pre-pregnancy weight (OR 0.98, CI 0.97–0.99 per kg), BMI (OR 0.95, CI 0.92–0.99 per kg/m2), and fasting OGTT at GDM diagnosis (OR 0.74, CI 0.59–0.93 per mmol/l) were associated with lower odds of postpartum testing. There was no association with practice characteristics and QOF outcomes.

In the multivariable analysis, higher parity (OR 0.54, CI 0.38–0.78), pre-pregnancy weight (OR 0.97, CI 0.95–0.99 per kg), gestational weight gain (OR 0.92, CI 0.87–0.98 per kg), and Index of Multiple Deprivation decile (meaning lower deprivation; OR 0.78, CI 0.65–0.95) were associated with lower odds of postpartum testing, adjusting for all other variables.

Consequences

In this population, 25% women with GDM did not attend postpartum testing. Attendance appears to be influenced by individual characteristics, such as parity and weight, rather than practice-related factors. In particular, women at highest risk of T2D are more likely to be those not attending testing and should be prioritised. Further work should focus on improving awareness and acceptability of T2D screening postpartum, particularly aiming to encourage more high-risk women to attend, therefore potentially improving long term outcomes.