Telemedicine for Insulin treated Gestational Diabetes Mellitus (TeleGDM): an exploratory randomised controlled trial of the effects of a web-based GDM support system on health service utilisation, maternal and foetal outcomes, costs and user experiences

Abstract

Gestational diabetes mellitus (GDM) is a condition characterised by elevated blood glucose that occurs in pregnancy and affects 11-15% of pregnant women. Tightly maintaining blood glucose levels (BGL) within target in GDM is associated with better maternal and foetal outcomes. In contrast poorly controlled hyperglycaemia is associated with adverse outcomes such as higher rates of caesarean delivery, macrosomia, foetal shoulder dystocia and admission of the new-born to a special care nursery or neonatal intensive care unit.

At least 50% of women with GDM need insulin to maintain glycaemic levels within target. In the early stages of insulin commencement, women with GDM often require intensive monitoring, frequent advice and support for insulin titration. This can be difficult to manage in pregnant women who are often faced with other competing demands, including work and caring for a young family, which contribute to the challenges of managing GDM. Further, the need for intensive support, together with inconsistent attendance for appointments, contribute to difficulties with managing GDM through the out-patient setting. Telemedicine has previously been shown to enhance patient self-monitoring and enabling provision of accurate and timely data transmission and sharing between patients and clinicians. Use of telemedicine potentially enabled efficient communication with timely response by clinicians in addressing urgent situations where BGLs are outside the desired target range. As a result, telemedicine may provide an innovative approach to streamline GDM management given the intensity of support and demand for insulin-treated GDM services and burden to GDM patients to attend appointments for ongoing monitoring and support.

My PhD study involved a complex intervention and as a result drew from elements of the three theoretical frameworks. First the Medical Research Council (MRC) Framework which guides the development and evaluation of complex interventions which draws attention to the steps to follow, i.e. intervention piloting and then proceeding to the exploratory phase which was the mainstay for my study. The second was the Normalisation Process Theory (NPT) which emphasises translating and embedding complex interventions in practice, noting the TeleGDM study was implemented in real practice. The third framework was the Telehealth Evaluation Framework which provides a guide to the elements to consider in order to standardise the evaluation of telehealth interventions.

My PhD project aimed to explore the impact of telemedicine on the management of insulin-treated GDM at Northern Health (NH). The project which is described in detail in this thesis consisted of: i) A background literature review including a systematic literature review and meta-analysis of telemedicine for GDM management; ii) Piloting the protocol at one of the campuses of NH. The pilot was aimed at gaining insights into the flow of the recruitment process and getting feedback on the chosen telemedicine system, before proceeding to an exploratory randomised controlled trial (RCT) stage; iii) An exploratory RCT comparing an adjunct telemedicine intervention to usual care in the management of insulin-treated GDM. The study primarily looked at impact on service utilisation, i.e. outpatient GDM clinic appointments. Other outcomes included a range of maternal and foetal/new-born clinical outcomes, patient and clinician satisfaction and service provider costs; and iv) A mixed methods outcome and process evaluation of the exploratory RCT. Ninety-four patients and five CDE-RNs participated in my study. The findings showed that telemedicine support in the management of GDM produced health service and clinical outcomes similar to usual care. Adjunct telemedicine support had no significant impact on the number of face-to-face outpatient GDM clinic appointments, foetal biometrics, rates of caesarean deliveries, macrosomia, large for gestational age, admission of new-borns to the special care nursery, birth-weight or costs. The intervention had the advantage of significantly reducing the time for patients to achieve optimum glycaemic control, an important outcome in GDM management. Importantly, while not superior to usual care in terms of health service use, telemedicine did not compromise the quality and safety of care in terms of foetal and maternal outcomes.

Uptake of the intervention, as reflected by the volume of GDM self-monitoring data entered into the telemedicine system by patients, showed that patients using this approach shared less data with clinicians, when compared to the usual care method of handwritten data. There was greater usage of the telemedicine system to share data in first four weeks of the intervention.

Statistically, there was no difference between the intervention and control on health service provider costs. However, limitations of scaling up the intervention notwithstanding, there was potentially for a significant cost saving from a health service perspective.

Themes from patient interviews showed that telemedicine as a concept, may be acceptable among patients, facilitated proactive self-management, and enabled personalised feedback. Some patients suggested telemedicine could potentially reduce face-to-face clinic attendances, thus, saving them travel time or allow them to balance work and ongoing GDM care. These views were possibly from a self-selected group of patients who engaged more with using the telemedicine system element of the intervention. Patients who engaged less with using the system were reluctant to participate in interviews.

As clinicians involved in performing the clinical aspects of the study, Credentialed Diabetes Education-Registered Nurses (CDE-RNs) had mixed responses regarding telemedicine. For instance, some expressed views that telemedicine may be a supplement usual care, rather than an alternative substitute. They also cited the telemedicine system’s technical design, and lack of integration with existing ehealth systems as the negatives of telemedicine, as these factors resulted in increased work to use the system, adversely impacting on workflow and productivity.

In conclusion, while my study was exploratory, telemedicine support for GDM showed no impact on service utilisation and provider costs. Telemedicine produced similar maternal and foetal clinical outcomes as usual care, suggesting no added risk to clinical quality of care, but with the possibility of a shorter time to insulin dose stabilisation. Further research in telemedicine using, user-friendly technological platforms that are fit for purpose, and including robust health economic evaluation in GDM is still needed.