The scale of the evidence behind this conclusion is itself striking. The researchers systematically searched five major databases, including PubMed, Embase, CINAHL, Web of Science, and the Cochrane Library, following the PRISMA reporting guidelines through April 2026. They included only cross-sectional studies that examined the relationship between the second victim phenomenon and either turnover intention or absenteeism among nurses. After independent screening, data extraction, and quality assessment using the Newcastle-Ottawa Scale, 38 studies made the final cut, encompassing 22,241 participants in the turnover intention analysis and 14,662 participants in the absenteeism analysis. That is one of the largest evidence bases ever assembled on this topic.

The statistical machinery behind the headline numbers deserves a closer look. Because the included studies varied in design, population, and measurement tools, the team used random-effects models to pool effect sizes, which explicitly account for variability between studies rather than assuming a single true effect. The pooled odds ratio for turnover intention came out at 2.99, with a 95 percent confidence interval of 2.81 to 3.18 and a p value below 0.001. In practical terms, a nurse experiencing second victim trauma had roughly triple the odds of intending to quit compared with a nurse who did not. For absenteeism, the pooled relative risk was 2.21, with a confidence interval of 2.05 to 2.37, meaning affected nurses were more than twice as likely to miss work.

One technical caveat that the authors confront head-on is heterogeneity, the statistical measure of how much the individual studies disagree with one another. The I-squared statistic, which expresses the percentage of variation in effect estimates attributable to study differences rather than chance, reached 75.9 percent for turnover intention and 82.6 percent for absenteeism. Both values indicate substantial heterogeneity, which is why the Cochran Q test confirmed significant variability and why the GRADE evaluation ultimately rated the quality of evidence as low, reflecting the observational design of the included studies. Yet the GRADE process also upgraded the magnitude of the effect to large, a judgment that the effect sizes themselves support: even the most conservative reading of the data places second victim trauma among the strongest known predictors of nurses wanting to leave.

To probe whether the association held across different contexts, the researchers performed subgroup analyses stratified by geographic region, healthcare setting, and study quality. The results were remarkably consistent: the effects were similar across varied geographic areas and healthcare environments, suggesting that second victim trauma is not a quirk of any particular health system or culture but a global occupational hazard. Publication bias, the risk that studies with null findings went unpublished and skewed the pooled estimates, was assessed with funnel plots and the Egger test, and neither showed significant asymmetry, with Egger test p values above 0.05 in both analyses. That means the tripled odds of quitting cannot be dismissed as an artifact of selective reporting.

The second victim phenomenon was first described in the medical literature more than two decades ago, when researchers documented that clinicians involved in patient safety events often suffer symptoms mirroring those of post-traumatic stress: intrusive memories, sleep disturbances, self-doubt, and a loss of confidence in their own competence. Nurses are particularly exposed because they spend the most time at the bedside, administer medications, and are frequently the first to discover that something has gone wrong. Previous research established that the phenomenon is common, but its quantitative implications for workforce retention and attendance had remained poorly understood, which is precisely the gap this meta-analysis was designed to fill.

The economic and human stakes of these numbers are difficult to overstate. Nursing turnover is one of the most expensive problems in healthcare, with each departing nurse costing hospitals tens of thousands of dollars in recruitment, orientation, and lost productivity, while the departure itself erodes team cohesion and institutional knowledge. Absenteeism compounds the damage: when nurses call in sick, their colleagues absorb the workload, overtime rises, and the risk of further errors climbs, potentially generating more second victims in a self-reinforcing cycle. With global health systems already facing projected shortfalls of millions of nurses by 2030, an occupational trauma that triples the odds of intending to quit represents a serious threat to workforce stability.

The authors argue that the findings point toward concrete organizational interventions rather than leaving nurses to cope alone. Peer support programs, in which trained colleagues provide structured emotional support after adverse events, psychological counseling services, and changes in institutional culture that replace blame with system-level learning are among the strategies highlighted. The underlying principle is that second victims are not incompetent practitioners but ordinary professionals reacting to extraordinary circumstances, and that organizations have a duty to support them just as they would support any worker injured on the job. Hospitals that have implemented such programs report faster recovery of affected staff and reduced intention to leave, though rigorous evaluation remains limited.

The study does carry limitations that shape how the results should be interpreted. All 38 included studies were cross-sectional, capturing a single moment in time, so they demonstrate association but not causation: it remains possible that nurses already inclined to leave are more likely to perceive and report second victim experiences. Heterogeneity was high, and the GRADE rating of low-quality evidence, driven by the observational design, means the estimates should be read as directional rather than precise. The authors call for longitudinal studies to examine temporal relationships and for intervention trials to test whether support programs actually reduce turnover and absenteeism, steps that would move the field from documenting the problem to solving it.

What makes this research resonate far beyond the nursing profession is the uncomfortable truth it reveals about modern healthcare: the people entrusted with saving lives can be quietly wounded by the work, and the wounds have measurable consequences for who stays and who walks away. As health systems worldwide battle burnout, staffing shortages, and rising attrition, this meta-analysis delivers a clear message to hospital administrators and policymakers alike. Supporting nurses through their darkest professional moments is not merely an act of compassion; it is a strategic imperative on which the stability of the entire healthcare workforce may depend.

Subject of Research: Second victim phenomenon and nurse turnover: systematic review and meta-analysis

Article Title: The impact of the second victim phenomenon on nurse turnover intentions and absenteeism: a systematic review and meta-analysis

Article References: Liu, J., Ma, D., He, J., Tan, D., Zhang, L., Peng, X., Sohaib, M., & Hu, S. (2026). The impact of the second victim phenomenon on nurse turnover intentions and absenteeism: a systematic review and meta-analysis. BMC Nursing, 25(1), Article 864. https://doi.org/10.1186/s12912-026-05247-x

Image Credits: AI Generated

DOI: 10.1186/s12912-026-05247-x

Keywords: impact, second, victim, phenomenon, nurse, turnover, intentions, absenteeism, systematic, review, meta-analysis, scientific research