Prevalence of Insomnia Symptoms in Older Chinese Adults During the COVID-19 Pandemic: A Meta-Analysis

Abstract

Background: The ongoing COVID-19 pandemic has disproportionately affected the sleep health of older adults, but the limited number of studies on insomnia symptoms of older Chinese adults differed in terms of screener of insomnia, sample size, and prevalence, making mental health planning for this population difficult. This meta-analysis estimated the prevalence of insomnia symptoms in older Chinese adults during the COVID-19 pandemic.Methods: Both Chinese (CNKI, Wanfang, VIP) and English (PubMed, EmBase, PsycInfo) databases were systematically searched to identify cross-sectional studies containing data on the prevalence of insomnia symptoms in older Chinese adults during the pandemic. Risk of bias (RoB) of included studies was assessed with the Joanna Briggs Institute Critical Appraisal Checklist for Studies Reporting Prevalence Data.Results: Nine studies with a total of 27,207 older Chinese adults were included. RoB scores of these studies ranged between zero and six. The pooled prevalence rates of insomnia symptoms and moderate and severe insomnia symptoms were 24.6% [95% confidence interval (CI): 19.5–30.5%] and 11.1% (95% CI: 7.2–16.9%), respectively. In subgroup analysis, significantly higher prevalence rates were observed in studies defining insomnia symptoms as “Insomnia Severity Index (ISI) ≥ 8” than in those defining them as “ISI ≥ 15” (32.6 vs. 15.6%, P < 0.001) and in older adults living in the COVID-19 epicenter than in those living in other places (35.2 vs. 23.3%, P = 0.006).Conclusion: Nearly one out of every four older Chinese adults suffered from insomnia symptoms during the pandemic. Mental health services for this population during the pandemic should include supportive activities aimed at improving mental well-being, periodic assessment of insomnia symptoms, and psychiatric assessment and treatment when necessary.

Anxiety symptoms in older Chinese adults in primary care settings: Prevalence and correlates

“…Fifth, some other factors associated with anxiety in older adults, such as personality and stressful life event, were not measured in this study. Finally, because our study was conducted before the COVID-19 pandemic and the pandemic has had a long-term and far-reaching negative impact on the mental health of older adults ( 19 , 68 ), studies during the pandemic and post-pandemic periods are warranted to further examine the influence of the pandemic on the risk of anxiety and its associated factors in older primary care attenders.…”

Clinical insomnia among elderly primary care attenders in Wuhan, China: A multicenter cross-sectional epidemiological study

“…In population-based studies using ISI and the same definition of clinical insomnia, the prevalence rates of clinical insomnia in rural community-residing older adults and older adults during the COVID-19 pandemic in China were 19.2 and 15.6%, respectively (22,35). Compared to these prevalence estimates in the elderly population, we found a much higher prevalence of clinical insomnia in Chinese EPCAs.…”

Prevalence of prolonged grief disorder and its symptoms in Chinese parents who lost their only child: A systematic review and meta-analysis

“…We used the Joanna Briggs Institute Critical Appraisal Checklist for Studies Reporting Prevalence Data (“JBI checklist” hereafter) to assess the level of RoB of included studies ( 21 ). The JBI checklist has nine items that are designed to assess three methodology domains of an individual study: sample representativeness (sample frame, sampling method, sample size, description of participants and setting), statistics (sample coverage of the data analysis, statistical analysis, response rate), and accuracy of the outcome assessment (validity of the instrument for assessing the outcome, standardization and reliability of the instrument for assessing outcome) ( 22 , 23 ). Each item has four answer options (yes, no, unclear, not applicable) and one point is allocated for a “yes” response, which yields a “0–9” total RoB score for a study with a higher score denoting lower RoB.…”

“…Prevalence data were synthesized by using the “metaprop” module of R, version 4.2.0, which adopted the one-step generalized linear mixed models with the logit link function ( 23 ). We used I2 statistic to test statistical heterogeneity between studies.…”