Use this URL to cite or link to this record in EThOS:
Title: The relationship between sleep and glucose control in gestational diabetes
Author: Alnaji, Alia Abdulahamid A.
ISNI:       0000 0004 6497 4277
Awarding Body: University of Leeds
Current Institution: University of Leeds
Date of Award: 2017
Availability of Full Text:
Access from EThOS:
Access from Institution:
This study set out to investigate the association between sleep among pregnant women with gestational diabetes (GDM) and their glucose control. Functional data analysis (FDA) methods were applied to glucose data collected via continuous glucose monitoring (CGM) systems. FDA is an advanced statistical method that respects the complexity of the dense auto-correlated data produced from repeated measurement of glucose over time. 192 pregnant women with GDM at their third trimester were recruited. Over a period of one week participants wore an actigraph (Actiwatch2 Respironics) which is a watch-like device on their non-dominant wrist to objectively measure their sleep, have a professional CGM system (iPro2 Medtronic) attached to them to continuously measure and record their interstitial glucose every 5 minutes, and complete the Pittsburgh Sleep Quality Index (PSQI) questionnaire to self-report their habitual sleep pattern for the previous month. Their demographic data and type of treatment they received were also collected. 152 participants had sufficient data retrieved from them, i.e. the PSQI questionnaire data and at least one night actigraphy-derived sleep data and one 24-hour day of CGM data. Using FDA methods, sequential glucose values data-points recorded over time with the CGM system were converted into a smooth 24-hour glucose curves with a functional form (as a function of time). The glucose curve was then used as one value, instead of the multiple data-points values it represents. Glucose control was assessed using the smooth glucose curves, as well as, a conventional summary metrics. The associations between participants’ actigraphy-derived and self-reported sleep characteristics and glucose control, were evaluated using standard and multilevel regression modelling for the conventional CGM data summary metrics and functional regression modelling for the smooth glucose curves. The study discovered a positive association between sleep disturbances and glucose control. Sleep disturbances were measured as poor sleep quality, short and long sleep durations compared to an average 6-8 hours sleep duration and difficulties in initiating and maintaining sleep. The timing and the amplitude of these associations were more apparent with FDA regression models than regression models with summary metrics. This study recommends the use of FDA in research involving the use of CGM systems, and encourages the clinician and the policy makers to consider sleep disturbances as a risk factor in glycaemic dysregulation in GDM.
Supervisor: Scott, Eleanor M. ; Ellison, George T. H. ; Law, Graham R. Sponsor: Ministry of Health ; Saudi Arabia
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available