Hip fractures are among the most consequential injuries of old age. As populations worldwide live longer, the incidence of these fractures continues to climb, and the consequences remain severe: one-year mortality after a geriatric hip fracture reaches twenty to thirty percent despite decades of progress in surgical technique, anesthesia, and perioperative care. Researchers have long concentrated on objective risk factors such as advanced age, male sex, heavy comorbidity burden, poor nutrition, and low preoperative functional status, all of which point to physiological frailty. What has been far less explored is the mind’s contribution to survival, and in particular fear of falling, a highly prevalent but frequently overlooked concern among older orthopedic patients.

The new study, led by Feng Gao, Yimin Chen, and Mingjian Bei of Beijing Jishuitan Hospital, Capital Medical University, together with colleagues, drew on a prospective observational cohort conducted at a tertiary care hospital in Beijing that operated a standardized orthogeriatric co-management model. Between November 2018 and November 2019, the team enrolled 1,092 patients aged sixty-five or older who underwent surgery for radiologically confirmed hip fractures of the femoral neck, intertrochanteric region, or subtrochanteric region, all admitted within twenty-one days of injury. Patients with cancer-related pathological fractures or periprosthetic fractures were excluded, as were those who did not provide informed consent.

Fear of falling was measured with the Chinese version of the Modified Falls Efficacy Scale, or mFES, an instrument originally developed by Tinetti and colleagues and later adapted by Hill and coworkers. The scale contains fourteen items, each scored from zero for no confidence to ten for complete confidence, with the total calculated as the average of all items and thus ranging from zero to ten. Lower scores indicate greater fear. Crucially, the mFES captures fall-related self-efficacy, a psychological construct reflecting a patient’s perceived confidence in performing daily activities without falling, which is distinct from objectively measured balance or physical performance. Because all participants were assessed within twenty-one days of injury, the authors argue that recall bias regarding pre-fracture functional confidence was minimized, supporting the validity of the preoperative measurement in an acute fracture population.

The cohort’s baseline profile reflected the typical fragility of a geriatric fracture population. The mean age was just over eighty years, and seventy-two percent of participants were women. The median mFES score was 8.04, with an interquartile range of 5.75 to 10.00. Hypertension was present in 57.1 percent of patients, diabetes mellitus in 27.7 percent, coronary heart disease in 22.3 percent, and prior stroke in 17.8 percent. More than a third of patients were classified as American Society of Anesthesiologists grade three or four, indicating serious systemic disease. Most patients, 70.1 percent, had walked independently before the fracture, while 28.1 percent needed walking aids. Femoral neck fractures accounted for just over half of cases, and cephalomedullary nailing was the most common operation, followed by hemiarthroplasty. The median interval from admission to surgery was about two days, and the median hospital stay just over five days.

Trained research nurses collected baseline data and ascertained survival status by telephone follow-up at four predefined time points after admission: thirty days, 120 days, one year, and three years. Complete follow-up data were available for 1,084 patients at thirty days, 1,074 at 120 days, 1,062 at one year, and 879 at three years, an attrition rate of roughly 19.5 percent at the final checkpoint. The primary outcomes were thirty-day postoperative mortality and all-cause mortality over the three-year follow-up period, analyzed with multivariable logistic regression for the short-term endpoint and Cox proportional hazards models for the long-term endpoint, adjusting for age, sex, ASA grade, stroke history, and pre-fracture mobility. The proportional hazards assumption was verified using Schoenfeld residuals, with no significant violation detected.

The findings were consistent and, in the short term, striking. In univariable logistic regression, higher mFES scores were significantly associated with lower mortality at thirty days, 120 days, and one year. After adjustment, the association remained statistically significant only at thirty days: each one-point increase in mFES score corresponded to a 22 percent reduction in the odds of dying within a month of surgery, with an adjusted odds ratio of 0.780 and a 95 percent confidence interval of 0.625 to 0.974. At 120 days and one year the adjusted associations fell just short of statistical significance, though a consistent trend toward lower mortality with higher self-efficacy persisted across every time point, strongest early and gradually attenuating thereafter.

Long-term survival told a similar but tempered story. In the univariable Cox analysis, higher mFES scores predicted a lower hazard of death over three years, and after multivariable adjustment the association endured, with an adjusted hazard ratio of 0.925, meaning each one-point gain in the scale corresponded to an approximately 7.5 percent reduction in the long-term hazard of death. The authors propose that the mechanisms differ between the two horizons. In the immediate perioperative phase, high fear of falling may amplify stress responses, suppress participation in early mobilization, undermine adherence to rehabilitation, and drive fear-related activity restriction, all of which raise the risk of early complications such as pneumonia, venous thromboembolism, delirium, and cardiopulmonary events. Low mFES scores are also closely tied to frailty, impaired mobility, and diminished functional reserve, all established predictors of poor early outcomes.

Over the longer term, the attenuated effect size suggests that fear of falling also reflects a broader, more persistent vulnerability. Patients with high fear often experience sustained reductions in physical activity, progressive functional decline, social isolation, and diminished engagement in rehabilitation and secondary prevention, factors that accumulate over years. Yet as follow-up lengthens, survival is increasingly shaped by competing forces, including chronic comorbidities, age-related physiological decline, malignancy, and socioeconomic determinants, which may dilute the relative contribution of any single preoperative psychological measure.

The clinical implications are immediate and actionable. The authors argue that preoperative fear-of-falling assessment deserves a place in routine orthogeriatric evaluation, since the mFES is brief, easy to administer, and capable of flagging psychologically vulnerable patients at heightened risk of postoperative death. Because fear of falling is potentially modifiable, early multidisciplinary interventions involving geriatricians, rehabilitation specialists, nursing staff, and psychological support services may improve perioperative management and recovery in this high-risk group. The study does have limitations: it was a post-hoc analysis of a single-center cohort, the mFES was applied only once at admission rather than the now-preferred Falls Efficacy Scale International, other psychological variables such as depression and anxiety were not systematically measured, and unmeasured factors like rehabilitation intensity and surgical technique may have influenced outcomes. Even so, the message is provocative and likely to resonate widely: in the fragile hours before hip fracture surgery, what an old and frightened mind believes about its own body may forecast survival as powerfully as any laboratory value, and a two-minute bedside questionnaire could help clinicians find the patients who need the most help the fastest.

Subject of Research: The association between preoperative fear of falling and postoperative mortality in older adults undergoing hip fracture surgery.

Article Title: Association between preoperative fear of falling and mortality after hip fracture surgery in older adults: a post-hoc analysis of a prospective cohort

Article References: Gao, F., Chen, Y., Bei, M., Chen, Y., Li, R., Xu, S., Chen, L., Tan, Z., Liu, G., Yang, M., & Wu, X. (2026). Association between preoperative fear of falling and mortality after hip fracture surgery in older adults: a post-hoc analysis of a prospective cohort. European Geriatric Medicine. https://doi.org/10.1007/s41999-026-01597-6

Image Credits: AI Generated

DOI: 10.1007/s41999-026-01597-6

Keywords: hip fracture, fear of falling, older adults, mortality, mFES, orthogeriatrics, geriatric surgery, fall-related self-efficacy, prognosis, frailty, rehabilitation, psychological vulnerability

The concept at the center of the framework is known as stress generation. Unlike stressors that occur independently of a person’s actions, stress-generating events are at least partly shaped by an individual’s behaviour, relationships or decisions. Examples might include escalating an argument, withdrawing from social contact, failing to meet an obligation because of low motivation, or repeatedly seeking reassurance in ways that strain relationships. The researchers emphasize that this process is not deliberate and should not be interpreted as blaming people for their suffering. Instead, stress generation describes a dynamic interaction between symptoms and the social environment, in which psychological difficulties can alter behaviour and those behavioural changes can produce new sources of stress.

The proposed model begins with changes in affective functioning, the systems involved in emotional experience and regulation. Internalizing disorders are commonly associated with heightened negative affect, reduced positive affect, emotional reactivity and difficulty returning to an emotional baseline after distress. A person may become more sensitive to rejection, more easily overwhelmed by ordinary setbacks or less able to experience reward from social and daily activities. These affective alterations can influence how situations are interpreted and how quickly a person responds to them. A minor disagreement, for example, may feel unusually threatening, while a neutral message from a friend may be experienced as evidence of disapproval. Over time, such emotional shifts can shape patterns of action that affect relationships, work and other areas of life.

The next stage involves everyday behaviour. When people are experiencing persistent sadness, fear, irritability or emotional exhaustion, they may avoid demanding situations, reduce communication, postpone responsibilities or respond more intensely during interpersonal conflict. These behaviours can provide short-term relief. Avoiding a difficult conversation may temporarily reduce anxiety, and withdrawing from social contact may protect someone from immediate embarrassment or perceived rejection. However, the same strategies can create longer-term consequences, including missed deadlines, financial complications, social isolation or resentment from family members and colleagues. In this way, behaviour that is understandable in the moment can unintentionally increase the probability of future stress.

The model gives particular importance to cognitive vulnerability, which can determine whether an emotional or behavioural reaction remains contained or develops into a larger chain of events. Cognitive vulnerability includes persistent patterns such as negative interpretations, rumination, hopeless expectations and beliefs that stressful outcomes are inevitable. Rumination can keep attention fixed on a conflict long after it has ended, while threat-focused thinking can encourage defensive or avoidant responses. When these cognitive processes interact with strong negative emotion, a person may repeatedly revisit the same event, interpret ambiguous actions in the most damaging way and respond as though a feared outcome has already occurred. That response may then provoke real interpersonal difficulties, providing apparent confirmation of the original belief.

This sequence can create a feedback loop between the individual and the surrounding social environment. An emotionally distressed person may withdraw, communicate less clearly or react defensively. Other people may respond with frustration, criticism or reduced support. The individual can then perceive those reactions as further evidence of rejection or hostility, increasing emotional distress and making additional problematic behaviour more likely. The process is not necessarily confined to one relationship or one type of stressor. Repeated cycles may spread across friendships, family interactions, romantic relationships, academic settings and workplaces. As stress accumulates, the resulting burden may worsen symptoms, creating the conditions for another round of stress generation.

The researchers describe this process as transdiagnostic, meaning that it may operate across multiple internalizing conditions rather than belonging to a single diagnosis. Depression, generalized anxiety, social anxiety and related forms of psychopathology often differ in their specific symptoms, but they can share underlying features such as negative affect, avoidance, interpersonal sensitivity and repetitive negative thinking. These common factors may help explain why stress generation is linked to impairment across diagnostic categories. Instead of treating each disorder as an isolated cause of stress, the framework maps how shared emotional, behavioural and cognitive mechanisms can produce similar patterns of escalating difficulty in different people.

A major implication of the model is that stress generation should not be viewed as a single event with a single cause. It is better understood as a process unfolding over time. Affective disruption may increase the likelihood of a particular behaviour, the behaviour may alter the social environment, and the resulting consequences may activate cognitive vulnerabilities that intensify emotional symptoms. The timing and sequence of these steps are important. A brief episode of avoidance may have little effect in one context but lead to substantial stress when it occurs repeatedly, affects an important responsibility or is interpreted by others as indifference. This process-based perspective may help researchers identify which links in the chain are most influential for different individuals.

The framework also points to several possible intervention targets. Treatment could focus directly on emotional regulation, helping people reduce reactivity and recover more effectively after distress. Behavioural approaches could address avoidance, withdrawal, disrupted routines and communication patterns that increase the likelihood of conflict or missed obligations. Cognitive interventions could target rumination, catastrophic interpretations and rigid expectations about rejection or failure. Because the model describes a reciprocal relationship between individuals and their environments, interventions might also include interpersonal strategies designed to repair damaged relationships, increase constructive communication and strengthen access to social support. Interrupting even one part of the cycle could potentially prevent later stressors from developing.

The Perspective does not present stress generation as an unavoidable consequence of internalizing psychopathology, nor does it suggest that individuals are responsible for every stressful event they experience. Rather, it offers a mechanistic account of how symptoms and circumstances can influence one another, sometimes producing self-reinforcing patterns that prolong illness and increase impairment. By bringing affective, behavioural and cognitive processes into a single framework, the model may help explain why some people recover after a stressful episode while others enter a cycle of accumulating difficulties. It also shifts attention toward prevention: identifying early changes in emotion, behaviour or thinking could make it possible to intervene before ordinary problems escalate into major life stress. For a field increasingly focused on personalized and transdiagnostic treatment, understanding how stress is generated may become a crucial step toward breaking the feedback loops that keep internalizing disorders alive.

Subject of Research: Stress generation processes in internalizing disorders

Article Title: A process model of stress generation in internalizing disorders

Article References: Starr, L.R., Dozois, D.J.A., Rnic, K. et al. A process model of stress generation in internalizing disorders. Nature Reviews Psychology (2026). https://doi.org/10.1038/s44159-026-00607-5

Image Credits: AI Generated

DOI: 10.1038/s44159-026-00607-5

Keywords: stress generation, internalizing disorders, depression, anxiety, affective functioning, cognitive vulnerability, behavioural processes, rumination, interpersonal stress, mental health interventions