Table 2.

Overview of studies involving monitoring of mental health, behaviour and glycaemic control

Study	Monitoring method	Sample	Key outcomes	Methodological characteristics
Mood and glycaemic control
Cox et al, 2007 [36]	SMBG	60 people with T1D	Postprandial excursions were associated with negative mood state and cognitive impairment	Observational Randomised Open label SMBG
Hermanns et al, 2007 [6]	CGM-blind	36 people with T1D	Higher glucose values were associated with less positive and more negative mood states Glycaemic variability showed no association with mood state	Observational Blinded CGM Multilevel analysis
Wagner et al, 2017 [37]	CGM-blind	50 people with T2D	Glycaemic variability had no association with mood state High and low glucose values were associated with negative affect	Observational Blinded CGM- Multilevel analysis
Skalf et al, 2009 [38]	SMBG	204 people with T2D	Negative mood predicted high fasting glucose the next day	Observational Open CGM Multilevel CGM
Shapira et al, 2021 [39]	SMBG	32 children / adolescents with T1D	Positive affect was associated with higher TIR, less time below range and less GV	Observational SMBG Multilevel analysis
Polonsky and Fortman, 2020 [40]	Open CGM	2019 people with T1D	Higher daily TIR was associated with better mood rating in the evening No association found between mood and GV	Observational Open CGM Multilevel analysis
Behaviour and glycaemic control
Wagner et al, 2017 [37]	Blind CGM EMA	50 people with T2D	Higher variability in self-care was associated with more hyper- and hypoglycaemic values	Observational Blinded CGM Multilevel analysis
Moscovich 2019, [47]	EMA	83 adults with T1D	Negative affect prior to meal was associated with more binge eating Binge eating was associated with higher postprandial glucose values	Observational Open CGM Multilevel analysis
Cecilia-Costa et al, 2021 [67]	Questionnaire	169 children / adolescents with T1D	Negative affect and higher diabetes distress were associated with more binge-eating episodes Disordered executive function was associated with more disordered eating behaviour	Observational SMBG or CGM
Yang et al, 2020 [49]	mHealth devices	60 people with T2D	Three phenotypes: low, medium and high engagement Low engagement was associated with higher HbA1c	Observational SMBG 6 month follow-up
Sleep and glycaemic control
Reutrakul et al, 2013 [68]	Sleep questionnaires	194 people with T2D	Lower sleep depth (<6 h) and unfavourable sleep chronotype were associated with higher HbA1c	Meta-analysis of observational studies Great heterogeneity
Knutson et al, 2011 [54]	Wrist actigraphy	40 people with T2D	Sleep fragmentation was associated with higher fasting glucose and higher HOMA index	Observational SMBG Multicentric

GV, glucose variability; T1D, type 1 diabetes; T2D, type 2 diabetes; TIR (time-in-range; glucose level 3.9–10 mmol/l)