Identification of the risk factors for insomnia in nurses with long COVID-19

Abstract

Purpose: To investigate the prevalence of insomnia among nurses with long COVID-19, analyze the potential risk factors and establish a nomogram model.

Methods: Nurses in Ningbo, China, were recruited for this study. General demographic information and insomnia, burnout, and stress assessment scores were collected through a face-to face questionnaire survey administered at a single center from March to May 2023. We used LASSO regression to identify potential factors contributing to insomnia. Then, a nomogram was plotted based on the model chosen to visualize the results and evaluated by receiver operating characteristic curves and calibration curves.

Results: A total of 437 nurses were recruited. 54% of the nurses had insomnia according to the Insomnia Severity Index (ISI) score. Eleven variables, including family structure, years of work experience, relaxation time, respiratory system sequelae, nervous system sequelae, others sequelae, attitudes toward COVID-19, sleep duration before infection, previous sleep problems, stress, and job burnout, were independently associated with insomnia. The R-squared value was 0.464, and the area under the curve was 0.866. The derived nomogram showed that neurological sequelae, stress, job burnout, sleep duration before infection, and previous sleep problems contributed the most to insomnia. The calibration curves showed significant agreement between the nomogram models and actual observations.

Conclusion: This study focused on insomnia among nurses with long COVID-19 and identified eleven risk factors related to nurses' insomnia. A nomogram model was established to illustrate and visualize these factors, which will be instrumental in future research for identifying nurses with insomnia amid pandemic normalization and may increase awareness of the health status of healthcare workers with long COVID-19.

Keywords: Insomnia; Long COVID-19; Nomogram; Nurses.

Fig. 1

Correlation factors among the nurses in insomnia

Fig. 2

An individualized nomogram model to describe the risk of insomnia in nurses Eleven variables were selected based on the results of the LASSO analysis. The R package “rms” was utilized to construct the prediction model of the post coronavirus insomnia nomogram. The factors with significant differences are indicated by asterisks. Points are assigned for each risk factor by drawing a line upward from the corresponding values to the ‘point’ line. The total points are the sum of the points obtained for the four risk factors and are plotted on the ‘total points’ line. The first row is taken as the observation data; that is, all risk factor points are calculated to have 563 risk scores corresponding to a risk of 3.33

Fig. 3

ROC curve of the nomogram model for assessing the quality of sleep in nurses with insomnia The ROC curve of the models, X-axis: specificity, Y-axis: sensitivity. The AUCs of the models were as follows: sleep score 0 (AUC = 0.892), sleep score 1 (AUC = 0.772), sleep score 2 (AUC = 0.865), and sleep score 3 (AUC = 0.974). This figure was drawn using R software version 4.2.1 http://www.R-project.org

Fig. 4

Calibration curve of the nomogram model for predicting insomnia in nurses The x-axis represents the probability of insomnia (ISI score ≧ 8), and the y-axis represents the actual probability. The 45-degree thick dotted line represents a perfect prediction. The thin dotted line represents the entire cohort (n = 437), and the solid line is bias-corrected by bootstrapping (B = 1000 repetitions), displaying the observed performance of the nomogram