Modifying the effects of sex on associations between suicidality and PTSD
The results of multinomial logistic regression to identify modifying effects after adjustment are graphically represented in figures 1 and 2. A notable sex-dependent variation in the association between suicidality and PTSD status emerged. Specifically, suicidality was significantly linked to PTSD in males but not in females, as indicated by significant interaction terms (figure 1 and online supplemental table S2). PTSD was diagnosed in 86 (8.2%) of 1047 participants at 3-month evaluation points, 73 (7.2%) of 1014 participants at 6-month evaluation points, 46 (4.7%) of 971 participants at 12-month evaluation points and 35 (3.8%) of 918 participants at 24-month evaluation points. Consistent with the overall findings, the relationship between suicidality and PTSD at each follow-up point remained significant solely in males with significant interaction terms (figure 2).
Figure 1Modifying effects of sex on the associations of suicidality at baseline with any post-traumatic stress disorder (PTSD) over 2 years in 1047 patients with physical injuries. ORs (95% CIs) were calculated for not present versus present suicidality at baseline on the development of any PTSD over 2 years adjusted for age, education, living alone, previous psychiatric disorders, previous traumatic events, any childhood abuse and scores on Hospital Anxiety and Depression Scale-anxiety subscale and Hospital Anxiety and Depression Scale-depression subscale. aSuicidality was evaluated by the Brief Psychiatric Rating Scale suicidality item score ranging from 1 (not present) to 7 (extremely severe). *p<0.05. ‡p<0.001.
Figure 2Modifying effects of sex on the associations of suicidality at baseline with post-traumatic stress disorder (PTSD) at 3, 6, 12 and 24 months in patients with physical injuries. ORs (95% CIs) were calculated for not present versus present suicidality at baseline on the development of any PTSD over 2 years adjusted for age, education, living alone, previous psychiatric disorders, previous traumatic events, any childhood abuse and scores on Hospital Anxiety and Depression Scale-anxiety subscale and Hospital Anxiety and Depression Scale-depression subscale. aSuicidality was evaluated by the Brief Psychiatric Rating Scale suicidality item score ranging from 1 (not present) to 7 (extremely severe). *p<0.05. †p<0.01. ‡p<0.001.
This 2-year longitudinal study in patients with physical injuries identified distinct sex-specific associations between suicidality and PTSD, notably revealing significant correlations in males. This correlation was consistent across all follow-up points (3, 6, 12 and 24 months), highlighting the importance of early suicidality in predicting PTSD, particularly in men.
Prior research has not systematically examined the sex differences in the predictive role of suicidality for PTSD in relation to the nature of traumatic events. This notable gap in the existing literature underscores the importance of our study, which is the inaugural exploration of these dynamics specifically within the context of patients who have experienced physical injury trauma. While focusing on this distinct group may somewhat limit the immediate generalisability of our findings, it crucially lays the groundwork for future research into how different types of traumas impact the complex interplay between suicidality and PTSD. Our study, therefore, provides pivotal initial insights and emphasises the need for comprehensive investigations across varied trauma contexts.
Several explanations are possible for the observed sex-specific associations. A primary consideration is the difference in help-seeking behaviour between the sexes. It is well documented that women are more likely to seek help for psychological problems, including suicidal ideation, often consulting general practitioners.5 6 In contrast, men may be more inclined to conceal their psychological struggles and less likely to seek professional help.11 This disparity in health-seeking behaviours could lead to differences in the reported incidence and diagnosis of PTSD. Second, research indicated that biological and neurological differences between sexes can influence the development and manifestation of PTSD and suicidality. Hormonal differences, particularly those involving oestrogen and testosterone, may affect how stress and trauma are processed in the brain. For instance, oestrogen has been linked to higher rates of PTSD in women due to its influence on the brain’s fear and stress response systems.12 Third, societal norms and expectations about gender roles can significantly impact how individuals experience and express distress. Men, for instance, may feel societal pressure to appear strong and unemotional, leading them to under-report symptoms of PTSD and suicidality.13 This could partly explain why the association between suicidality and PTSD is observed more in men, as they might only express these symptoms when they are particularly severe.
The limitations of this study include the recruitment of participants exclusively from a single trauma centre, which, while beneficial for consistency in evaluation and follow-up, may limit the generalisability of the results. Suicidality was assessed with the BPRS item, effective for evaluating ideation and behaviour severity.14 However, this single item might not fully capture the complexity of suicidal thoughts and behaviours. Although follow-up evaluations were conducted via telephone interviews, a method shown to be as valid as face-to-face interviews in previous research,10 it is essential to consider the potential limitations of this approach.
A major strength of our study was the methodical recruitment of all eligible patients who had recently experienced physical injuries, minimising selection bias and ensuring a representative sample. We employed a structured research protocol with frequent follow-up assessments to mitigate bias from variable examination times. Consistency and reliability were further ensured by using standardised scales like the CAPS-5 for PTSD diagnosis. Additionally, we collected a wide range of covariates at baseline, and our long-term follow-up showed no evidence of selective attrition, bolstering the study’s robustness.
In conclusion, this study provides significant insights into the sex-specific associations between suicidality and PTSD following physical injuries, emphasising the crucial role of gender in PTSD development. Our results advocate for gender-sensitive public health strategies to prevent PTSD and suicidality and call for clinicians to tailor treatments to the distinct experiences of men and women. Highlighting the need for early, comprehensive suicidality screening, this study points towards the necessity of timely mental health interventions. Future research should expand to multicentre studies and various trauma types. Moreover, investigating the relationships between fluctuations in suicidality over time and PTSD is worthwhile. Such inquiries are vital for advancing our understanding of these complex relationships, ultimately guiding more effective prevention and treatment methods.
