Table 3.

Summary of methodological characteristics of the review studies.

Domain	Number (%)	Studies
Study design
Cohort study	13 (56.5)	(8, 10, 12–14, 18, 39–44)
Cross-sectional	10 (43.5)	(5, 17, 45–52)
Sample characteristics
Studies inclusion of both sexes	18 (78.2	(5, 8, 10, 12–14, 40–48, 50–52)
Studies included only male	4 (17.4)	(11, 17, 18, 49)
Studies consisted of male participants (male representation 60% or more)	5 (21.7)	(8, 10, 13, 40, 47)
Methodology characteristics
Sample power calculation included	3 (13.0)	(5, 8, 13)
Comparator
Matched control	2 (8.7)	(40, 43)
Unmatched control	21 (91.3)	(5, 8, 10–14, 17, 18, 39, 41, 42, 44–52)
Direction of association
OSA increases the risk of T2DM	10 (43.5)	(8, 10–14, 18, 39, 41, 42).
T2DM increases the risk of OSA	1 (4.3)	(43)
Bidirectional association of OSA and T2DM	2 (8.6)	(40, 44)
Unspecified direction	10 (43.5)	(5, 17, 45–52)
Methods of assessment for OSA
Objective OSA assessment (e.g., PSG or home portable sleep monitors)	15 (65.2)	(8, 10–14, 17, 18, 41, 47–52)
Subjective OSA assessment (e.g., sleep questionnaire)	5 (21.7)	(5, 39, 44–46)
Medical record codes for OSA	3 (13.0)	(40, 42, 43)
Standard objective assessment OSA scoring criteria	9 (39.1)	(8, 10, 13, 14, 41, 49–52)
Standard subjective assessment OSA scoring criteria	2 (6.8)	(46, 49)
Methods of assessment for T2DM
Objective T2DM assessment (e.g., HbA1C)	17 (73.9)	(5, 8, 10–12, 14, 17, 18, 39, 45–52)
Subjective T2DM assessment (e.g., self-report)	2 (8.6)	(41, 44)
Medical record codes for T2DM	4 (17.4)	(13, 40, 42, 43)
Other methodological aspects
Missing data treatment addressed	11 (47.8)	(5, 10, 13, 17, 39, 42, 43, 48–50, 52)
Attrition rate reported in cohort studies	6 (46)	(10, 12–14, 18, 41)
Sensitivity analysis conducted	10 (43)	(5, 8, 10, 13, 14, 39–41, 47, 52)
Adequate report of limitations and residual factors	13 (56.5)	(10–13, 17, 18, 41, 42, 45–47, 49, 50).

OSA, obstructive sleep apnea; T2DM, type 2 diabetes mellitus; PSG, polysomnography; HbA1C, haemoglobin A1C.