Discrepancy between objective and subjective measurements of sleep quality: the role of panic-agoraphobic spectrum

INTRODUCTION: There is evidence that anxiety and depressive symptoms may lead individuals to under-estimate their own sleep quality, particularly among younger subjects (aged <45 yrs). OBJECTIVES: The aim of this study was to investigate the discrepancy between objective and subjective measurements of sleep quality in a sample of healthy control subjects (HCs) with no Axis I mental disorders, and the possible impact of panic-agoraphobic spectrum symptoms. METHODS: A total of 117 HCs (65 males and 97 females; Age: 35.3±14yrs) were evaluated by the: Panic Agoraphobic Spectrum-Self Report (PAS-SR), to investigate panic spectrum; the Pittsburgh Sleep Quality Index (PSQI) and actigraphy, respectively for the subjective and the objective sleep efficiency (SE) measures. Groups were divided according to the congruence between SE-actigraphic vs SE-PSQI (“Accurate”, “Underestimate”, “Overestimate”), establishing as a threshold an SE>85% as a measure of good SE. Regression analyses were conducted to assess the association between PAS-SR domains and the discrepancy between objective and subjective measurements, controlling confounding factors such as age, gender and BMI RESULTS: Since our data showed that a low sleep quality was associated with a greater age and that higher PAS-SR scores were associated with younger age, we used a sub-sample of 117 participants with age <45 years and comparing the 3 groups of subjects created on the basis of the discrepancy: Accurate, N = 74 (63.2 %), “Overestimate group”, N= 23 (19.7 %), “Underestimate group”: N=20 (17.1 %), we found a statistically significant difference among groups in the PAS.SR separation anxiety domain (p value=0.032), with a multinomial regression model confirming this domain contributed significantly to the differentiation between the three groups with higher symptoms being associated with a higher probability of belonging to the “underestimate” group. CONCLUSIONS: Our results suggest that the discrepancy between objective and subjective sleep efficiency measurements in HCs could be affected by panic spectrum symptoms, particularly separation anxiety. DISCLOSURE OF INTEREST: None Declared