Review question
To assess the effects of reminder systems to increase uptake of testing for type 2 diabetes or impaired glucose tolerance in women with a history of gestational diabetes mellitus (GDM).
Background
Some women experience high blood glucose concentrations during pregnancy (termed GDM). Although these high blood glucose concentrations usually normalise immediately after birth, women who have experienced GDM are at an increased risk of developing type 2 diabetes in the future. It is therefore important that they are regularly tested for higher than normal blood glucose levels (to detect type 2 diabetes or 'impaired glucose tolerance' which is a prediabetic state sometimes preceding type 2 diabetes), starting in the months after they have given birth. However, for a variety of reasons, many women do not get their blood glucose tested after experiencing GDM.
Study characteristics
A single study of 256 women who had experienced GDM whether posting reminder letters to 213 women or their doctors, three months after the birth of a baby, would help to increase the number of women taking a blood glucose test compared with 43 women sent no reminder.
Key results
This study showed that, compared with no reminder, a postal reminder was around two to four times (depending on the blood glucose test concerned) more likely to encourage women who had experienced GDM to take a blood glucose test three months after having their baby. It did not seem to make a difference if the reminder was sent to the woman only, the physician only or to both the woman and the physician.
The trial did not assess women's quality of life, or how many women were subsequently diagnosed with type 2 diabetes or impaired glucose results after giving birth.
Other kinds of reminders such as email and telephone need to be assessed in studies as they might be easier and more convenient for women than posted reminders. We need to know more about women's preferences and attitudes, and also to find out whether increasing the chances of a woman being tested helps to reduce her risk of developing type 2 diabetes in the future, for example by encouraging a healthier diet and more exercise.
Quality of the evidence
The overall quality of evidence was considered low as the only included study involved few numbers of participants and provided imprecise results.
Currentness of data
This evidence is up to date as of June 2013.
Results from the only trial that fulfilled our inclusion criteria showed low-quality evidence for a marked increase in the uptake of testing for type 2 diabetes in women with previous GDM following the issue of postal reminders. The effects of other forms of reminder systems need to be assessed to see whether test uptake also increases when email and telephone reminders are deployed. We also need a better understanding of why some women fail to take opportunities to be screened postpartum. As the ultimate aim of increasing postpartum testing is to prevent the subsequent development of type 2 diabetes, it is important to determine whether increased test uptake rates also increase women's use of preventive strategies such as lifestyle modifications.
The early postpartum period is an important time in which to identify the risk of diabetes in women with a history of gestational diabetes mellitus (GDM). Oral glucose tolerance and other tests can help guide lifestyle management and monitoring to reduce the future risk of type 2 diabetes mellitus.
To assess whether reminder systems increase the uptake of testing for type 2 diabetes or impaired glucose tolerance in women with a history of GDM.
We searched MEDLINE and EMBASE (last searched 1 June 2013) and The Cochrane Library (last searched April 2013).
We included randomised trials of women who had experienced GDM in the index pregnancy and who were then sent any modality of reminder (or control) to complete a test for type 2 diabetes after giving birth.
Two authors independently screened titles and abstracts for relevance. One author extracted the data, carried out 'Risk of bias' assessments and evaluated the overall study quality according to GRADE (Grading of Recommendations Assessment, Development and Evaluation) criteria; the other author double-checked these procedures. Meta-analysis was not possible as only one study was eligible for inclusion.
Only one trial with an unclear risk of bias in the majority of domains was included in the study; the overall study quality was judged to be low. This factorial trial of 256 women compared three types of postal reminder strategies (in a total of 213 women) with usual care (no postal reminder, 43 women) and reported on the uptake of four possible types of glucose tests. The three strategies investigated were: reminders sent to both the woman and the physician; reminder sent to the woman only; and reminder sent to the physician only, all issued approximately three months after the woman had given birth.
There was low-quality evidence that all three reminder interventions increased uptake of oral glucose tolerance tests compared with usual care (no reminder system): reminders to the woman and the physician (uptake 60% versus 14%): risk ratio 4.23 (95% confidence interval (CI) 1.85 to 9.71); 116 participants); reminder to the woman only (uptake 55% versus 14%): RR 3.87 (95% CI 1.68 to 8.93); 111 participants); reminder to the physician only (uptake 52% versus 14%): RR 3.61 (95% CI 1.50 to 8.71); 66 participants). This represented an increase in uptake from 14% in the no reminder group to 57% across the three reminder groups. There was also an increase in uptake of fasting glucose tests in the reminder group compared with the usual care group: reminders to the woman and the physician versus no reminder (uptake 63% versus 40%): RR 1.57 (95% CI 1.01 to 2.44); reminder to the woman only (uptake 71% versus 40%): RR 1.78 (95% CI 1.16 to 2.73); reminder to the physician only (uptake 68% versus 40%): RR 1.69 (95% CI 1.06 to 2.72). Uptake of random glucose and glycated haemoglobin A1c tests was low, and no statistically significant differences were seen between the reminder and no reminder groups for these tests. Uptake of any test was higher in each of the reminder groups compared with the no reminder group (RR 1.65 (95% CI 1.12 to 2.41); 1.73 (95% CI 1.18 to 2.52); and 1.55 (95% CI 1.01 to 2.38) in the respective reminder groups.
The trial did not report this review's other primary outcomes (proportion of women diagnosed with type 2 diabetes or showing impaired glucose tolerance or impaired fasting glucose after giving birth; or health-related quality of life). Nor did it report any secondary review outcomes such as diabetes-associated morbidity, lifestyle changes, need for insulin, recurrence of GDM or women's and/or health professionals' views of the intervention. No adverse events of the intervention were reported.
Subgroup interaction tests gave no indication that dual reminders (to both women and physicians) were more successful than single reminders to either women or physicians alone. It was also not clear if test uptakes between women in the reminder and no reminder groups differed by type of glucose test undertaken.