INTRODUCTION
Zolpidem is one of the most frequently prescribed medications for psychiatric outpatients managing insomnia [1]. While it is widely recognized for its rapid and potent hypnotic effects, concerns have been raised regarding its potential adverse effects, including fatigue, cognitive impairment, falls, and infections [2]. Beyond its physical side effects, increasing attention has been directed toward its psychiatric side effects [3]. Insomnia, the most prevalent sleep disorder [4], is frequently comorbid with psychiatric disorders, a relationship that has been extensively studied [5]. Among psychiatric concerns, the association of zolpidem use and suicidality—a phenomenon with profound clinical implications—has garnered significant interest [6-8]. Recent meta-analyses have highlighted a significantly increased risk of suicide, suicide attempts, and suicide-related mortality associated with zolpidem use [9]. However, a recent population-based retrospective cohort study reported no significant association of zolpidem use and suicidality when used for less than 80 months, with a notable correlation observed only when usage extended beyond 80 months [10].
These findings necessitate careful interpretation, particularly given the high co-occurrence of insomnia and psychiatric disorders. Established psychiatric factors, such as the severity of depression and insomnia, are well-documented contributors to suicide risk [11,12]. Additionally, parasomnia—a known side effect of zolpidem [3]—has been linked to self-injurious behavior [13,14]. Thus, accounting for diverse psychiatric variables is crucial when investigating the association of zolpidem use and suicidality. This study aims to investigate whether zolpidem use is independently associated with suicidality after adjusting for confounding variables, including psychiatric symptom severity and demographic factors. Moreover, given the widespread prescription of zolpidem among psychiatric outpatients [15], its effects within this specific population warrant detailed examination.
The hypotheses of this study are as follows: 1) insomnia patients using zolpidem will exhibit statistically significantly higher levels of suicidality compared to those not using zolpidem, and 2) even after controlling for demographic differences and psychiatric symptom severity, the zolpidem use group will demonstrate statistically significantly higher levels of suicidality.
METHODS
Data collection
This study utilized data obtained from the medical records of 6,193 patients (mean age: 48.96 years; 38.4% male, 61.6% female) who attended the Department of Psychiatry at Samsung Medical Center, a tertiary hospital located in Seoul, South Korea, between January 2018 and December 2022. Of these patients, 3,128 individuals presenting with complaints of insomnia were included in the analysis. Insomnia symptoms were assessed through a comprehensive evaluation encompassing psychiatric history, mental status examination, and psychological testing. Diagnoses of depressive disorders were established by licensed clinical psychologists through structured clinical interviews, using either the Mini International Neuropsychiatric Interview-Plus based on the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV), or the Structured Clinical Interview for the DSM-5, depending on the version in use at the time of assessment. Medication usage data—including antidepressants (selective serotonin reuptake inhibitors, serotonin-norepinephrine reuptake inhibitors, and others), anxiolytics (benzodiazepines), antipsychotics (typical and atypical antipsychotics), and mood stabilizers (lithium, valproate, lamotrigine, carbamazepine, and others)—were also extracted. Suicidality was determined based on the presence of any of the following: current suicidal ideation, current suicidal plans, or current/past suicide attempts. Suicidality assessments were conducted by six mental health professionals, with unclear cases subjected to further review by a board-certified psychiatrist. Ethical approval for the study was granted by the Institutional Review Board of Samsung Medical Center (IRB No. 2020-11-107) which constituted a waiver of the requirement for written informed consent.
Zolpidem use
Among the insomnia cohort, patients with documented use of zolpidem in medical records from external facilities or a history of zolpidem prescriptions were categorized into the zolpidem use group.
Severity of depression and insomnia
Depression severity was assessed using the Hamilton Depression Rating Scale (HAM-D), a clinician-administered tool that evaluates depressive symptoms experienced over the previous week through a semi-structured interview [16]. The Korean version of the HAM-D, adapted by Yi et al. [17], was utilized. The 17-item scale yields a total score ranging from 0 to 52, with a Cronbach’s alpha value of 0.77 (95% confidence interval [CI] 0.76-0.78) in this study, indicating reliable internal consistency. Insomnia severity was evaluated using items 4, 5, and 6 of the HAM-D, which assess initial, middle, and terminal insomnia. The remaining items (1-3 and 7-17) were summed to quantify depression severity.
Statistical analysis
Group differences in distributions or values were assessed using chi-square tests and independent sample t-tests, with results reported as χ2 or t statistics and corresponding p-values. Logistic regression analysis was conducted to examine the association of zolpidem use and suicidality, while adjusting for potential confounders. The results of the regression analysis are presented with β (estimates), standard errors, zvalues, odds ratio, 95% CI, and p-values. Statistical significance was defined as p<0.05 for all analyses. Data were analyzed using SPSS version 29.0 (IBM Corp.).
RESULTS
Characteristics of respondents by zolpidem usage
Among the 3,128 patients reporting insomnia, 364 were zolpidem users. Compared with the 2,764 individuals in the control group who reported insomnia but did not use zolpidem, zolpidem users were more likely to be female (69.8% vs. 62.6%; p=0.009), older (mean age: 55.77 vs. 48.60; p<0.001), and married (62.5% vs. 47.1%; p<0.001). Among zolpidem users, the prevalence of concomitant antidepressants (58.5% vs. 47.1%; p<0.001) or anxiolytics (60.7% vs. 46.9%; p<0.001) use was higher. Furthermore, zolpidem users demonstrated more severe insomnia, as indicated by higher HAM-D insomnia subscores (3.24 vs. 2.75; p<0.001). No significant differences were observed between the groups with respect to occupational status, education level, antipsychotics use, mood-stabilizers use, depressive disorder diagnosis, and depression severity (Table 1). Statistical comparisons were performed using chi-square tests or independent sample t-tests, as appropriate.
The association of zolpidem use and suicidal tendencies
The association of zolpidem use and suicidality is summarized in Table 2, where comparisons between the zolpidem and control groups were conducted using chi-square tests. Suicidal tendencies, same as suicidality, were defined by the presence of any of the following: suicidal ideation, suicidal planning, or suicidal attempt. Zolpidem users exhibited a higher prevalence of suicidal tendencies (44.8% vs. 39.3%; p=0.046) and suicide attempts (23.6% vs. 17.4%; p=0.004). However, no statistically significant differences were identified for suicidal ideation (40.1% vs. 35.6%; p=0.095) or suicidal planning (8.0% vs. 7.6%; p=0.836).
Logistic regression analysis of zolpidem use and suicidal tendencies
Logistic regression analysis was conducted to evaluate the association of zolpidem use and suicidality while adjusting for potential confounding variables, including gender, age, marital status, antidepressants use, anxiolytics use, depression severity, and insomnia severity (Table 3). The analysis revealed that zolpidem use was significantly associated with increased suicidal tendencies (β=0.314, standard error=0.137, odds ratio=1.368, 95% CI 1.05-1.79, p=0.022), suicidal ideation (β=0.326, standard error=0.138, odds ratio=1.386, 95% CI 1.06-1.82, p=0.018), and suicide attempts (β=0.379, standard error=0.157, odds ratio=1.460, 95% CI 1.07-1.98, p=0.016). No significant association was found between zolpidem use and suicidal planning, even after adjustment for confounders (β=0.207, standard error=0.227, odds ratio=1.231, 95% CI 0.79-1.92, p=0.360).
DISCUSSION
The study investigates the association of zolpidem use and suicidality, accounting for psychiatric factors that could influence this association. Given the high prevalence of insomnia and its frequent comorbidity with psychiatric disorders, the study focused on psychiatric outpatients presenting with insomnia. Within this cohort, comparisons were made between patients who used zolpidem and those who did not. After adjusting for insomnia severity, depression severity, and demographic factors, the zolpidem group demonstrated higher levels of suicidal ideation and suicide attempts. However, no significant difference in suicidal planning was observed between the groups.
The lack of a statistically significant difference in depression severity between the zolpidem and control groups underscores the notable association of suicidal ideation and zolpidem use. Prior studies have highlighted that insomnia may influence suicidality independently of depression [18]. This study separates these factors and incorporates appropriate statistical adjustments. Although zolpidem use has been shown to improve sleep, it does not alleviate symptoms of anxiety or depression [19]. Moreover, zolpidem use has been linked to impairments in cognitive function [20,21] and emotional regulation [22]. These combined effects may contribute to the observed association of zolpidem use and suicidal ideation, independent of depression severity, consistent with conclusions from prior meta-analyses [9].
While the zolpidem group exhibited higher levels of suicidal ideation, this finding was not associated with increased suicidal planning but rather was more closely linked to suicide attempts. Recent evidence suggests that zolpidem use is associated with an elevated risk of suicide attempts in the short term, whereas no significant relationship has been observed with suicide attempts during long-term use [23]. The acute suicidal risk linked to zolpidem and its side effects, such as parasomnia related compulsive night eating and cravings [24,25], raises questions about its potential relationship with impulsivity. Although a definitive link between zolpidem and impulsivity has not yet been established, case series have reported instances suggestive of such a connection [26]. This context may explain why zolpidem is associated with suicide attempts rather than suicidal planning, aligning with prior studies linking zolpidem to suicidal behavior [27].
Notably, the statistical significance of the association of zolpidem use and suicidality increased markedly after adjusting for insomnia, antidepressants use, anxiolytics use, and depression severity. For example, p-values improved for suicidal tendencies (from p=0.046 to p=0.022), suicidal ideation (from p=0.095 to p=0.018), and suicidal planning (from p=0.836 to p=0.360). This finding suggests that psychiatric comorbidities, such as depression, anxiety, and insomnia, may have a masking effect on the association of zolpidem use and suicidality. Since zolpidem may be more frequently prescribed to patients with milder depression and lower suicidal risk, its relationship with suicidality might be less apparent in clinical settings [28,29]. However, the present study highlights that the association of zolpidem use and suicidality becomes more evident among psychiatric outpatients with comparable levels of insomnia or depression. Moreover, given that the association of zolpidem use and suicidality persists even after controlling for the use of antidepressants and anxiolytics—and considering the frequent co-prescription of these medications—caution is warranted when prescribing zolpidem to patients with mood disorders.
A key strength of this study is its utilization of data collected directly from the patient population, enabling the inclusion of clinically relevant variables such as depression severity and insomnia severity in the analysis. This methodology enhances the applicability of the findings to real-world clinical settings, particularly those involving patients with insomnia and depression for whom zolpidem is routinely prescribed. Despite the strengths of this study, several limitations should be acknowledged, which may restrict the generalizability and interpretability of the findings. First, the use of hospital-based, retrospective medical record data limits the ability to establish a clear temporal or causal association of zolpidem use and suicidality. Second, the lack of detailed information regarding the dosage, duration, and cumulative exposure to zolpidem precludes a more nuanced analysis of dose-response effects or long-term outcomes. Furthermore, although basic clinical data were available, the absence of standardized psychiatric diagnostic procedures and structured psychological assessments—such as validated diagnostic interviews or comprehensive psychometric testing—limits the ability to control for underlying psychiatric comorbidities and symptom profiles that may influence both zolpidem use and suicidality. These methodological constraints should be carefully considered when interpreting the study’s conclusions.
In conclusion, this study suggests that zolpidem use in patients with insomnia may be associated with increased suicidality, even after adjusting for the severity of depression and insomnia, as reflected in higher rates of suicidal ideation and attempts. Considering zolpidem’s widespread use for insomnia, it is imperative to monitor patients for suicidality during treatment. Clinicians should carefully weigh the risk-benefit profile of zolpidem and individual patient characteristics when prescribing this medication. Further research is required to confirm these findings in similar populations, with particular attention to the duration of treatment and cumulative dosage of zolpidem.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Print
