Article
Intake of sleep disturbing and sleep inducing drugs – an interaction analysis on the additive scale.
Search Medline for
Authors
-
Anna-Therese Lehnich
- IMIBE, Universität Duisburg-Essen, Essen, Germany
-
Bernd Kowall
- IMIBE, Universität Duisburg-Essen, Essen, Germany
-
Susanne Moebus
- IMIBE, Universität Duisburg-Essen, Essen, Germany
-
Karl-Heinz Jöckel
- IMIBE, Universität Duisburg-Essen, Essen, Germany
-
Andreas Stang
- IMIBE, Universität Duisburg-Essen, Essen, Germany
| Published: | December 5, 2017 |
|---|
Outline
Text
Background: Drug-drug-interactions (DDIs) are a growing public health concern in ageing populations suffering from comorbidities. An epidemiologic method to identify interaction in biologic contexts is to estimate interaction contrasts (ICs). We analyzed the intake of drugs with sleep disturbing and sleep inducing side effects and the association with self-reported sleep characteristics, applying ICs.
Materials and Methods: We used data from the baseline examination (2000-2003, n=4,814, 49.8% male, 45-75 years old) of the population-based Heinz Nixdorf Recall Study, Germany. The interview provided information on sleep disturbances, nocturnal sleep duration, siesta and drug intake. From 3,314 participants with drug intake but without intake of hypnotics, we created four disjoint categories: intake of drugs not affecting sleep (D̅ I̅), sleep disturbing drugs (DI̅), sleep inducing drugs (D̅I) and intake of both sleep disturbing and inducing drugs (DI). We estimated prevalence differences (PDs) with 95% confidence limits (95% CI) via stratified linear regression models. According to a directed acyclic graph we adjusted for age, several diseases, alcohol intake and education level. We calculated ICs as differences of adjusted PDs.
Results: We found little interaction between sleep disturbing and sleep inducing drugs concerning sleep disturbances. However, the adjusted PDs for siesta in women were -0.07 (95% CI: -0.16;0.02) and 0.13 (95% CI: 0.06;0.20) for the difference between categories DI̅ and D̅ I̅ and the difference between categories DI and D̅I, respectively, resulting in an IC of 0.19 (95% CI: 0.08;0.31). The corresponding adjusted PDs for siesta in men were 0.06 (95% CI: -0.04;0.16) and -0.00 (95% CI: -0.08;0.08), resulting in an IC of -0.06 (95% CI: -0.19;0.07).
Conclusion: Interaction effects in relation to siesta qualitatively differed between men and women. This might also be due to the intake of different drugs in men and women. Our analysis suffered from low statistical power.
