TABLE 1

Characteristics of published studies that investigated a longitudinal association between obstructive sleep apnoea (OSA) and depression

Peppard et al., 2006 [10] Wisconsin Sleep CohortChen et al., 2013 [11] Health Insurance Database, TaiwanLang et al., 2015 [12] Australian Community-based cohort (abstract)
Subjects n1408OSA 2818; control 14 0901600
Male %5667100
Age yearsMean (range) 51 (30–75)Mean±sd 46.4±15.7Range 35–83
Definition of depressionModified Zung
depression scale ≥50
or antidepressant use
Excluded: “I have trouble sleeping through the night” and “I get tired for no reason”
Incident diagnosis of DD
ICD-9 codes:
296.2, 296.3, 300.4 and 311
BDI/CES-D
No details
Definition of OSAFull PSG
AHI categories: 0–<5;
5–<15; ≥15
ICD-9 codes: 327.23, 780.51, 780.53 or 780.57 following PSGHome PSG
Follow-up4 years
n=259 (transitioning from not depressed to depressed, or vice versa)
1 year
n=167 (received a diagnosis of DD: inpatient or outpatient)
5 years
Not reported (small given wide confidence interval)
AnalysesGEEM
Conditional logistic regression
Model: age, BMI, alcoholic intake and history of CVD
Cox regression
Model: sex, age, urbanisation level, index date, hypertension, diabetes, CHD, hyperlipidaemia, obesity and alcohol abuse/dependency
Logistic regression (?)
Model: no details
EffectIncrease to the next higher category
OR 1.8 (95% CI 1.3–2.6)
OSA (yes/no)
HR 2.18 (95% CI 1.55–3.08)
Severe OSA (AHI ≥30)
OR 2.9 (95% CI 1.19–6.92)
  • DD: depressive disorder; ICD: International Classification of Diseases; BDI: Beck depression inventory; CES-D: Centre for Epidemiologic Studies depression scale; PSG: polysomnography; AHI: apnoea–hypopnoea index; GEEM: generalised estimating equation model; BMI: body mass index; CVD: cardiovascular disease; CHD: coronary heart disease; OR: odds ratio; HR: hazard ratio.