The research team, led by Behire Sançar and colleagues, defines bad news in the clinical context as information about a serious diagnosis, an unfavorable turn in treatment, or any revelation that negatively alters how patients envision their future. Communication of this kind is not merely a courtesy; it is an ethical obligation with measurable consequences for psychological adjustment, trust in care teams, and downstream health outcomes. Nurses occupy a unique position in this process because they have the closest and most sustained contact with patients, meaning they often absorb the first wave of fear, grief, and questioning that follows a diagnosis. Previous studies have repeatedly shown that nurses feel poorly prepared for these encounters, reporting inadequate training, difficulty choosing appropriate communication methods, and difficulty carrying the emotional burden, which in turn contributes to occupational stress and burnout.

To quantify perceived knowledge and skill, the researchers administered the Bad News Delivering Scale for Health Professionals, a validated instrument developed by González-Cabrera and colleagues and adapted into Turkish by Karataş Baran and colleagues. The scale comprises items scored on a three-point Likert range from “never” to “always,” yielding total scores between 21 and 63, with higher values indicating greater perceived ability. The instrument demonstrated excellent internal consistency in this sample, with a Cronbach’s alpha of 0.94, compared with 0.87 in the Turkish adaptation study. Participants also completed a 12-item descriptive form covering demographics, clinical unit, self-assessed ability, and history of negative experiences. Data were collected face-to-face in November and December 2024, each interview lasting roughly 20 minutes, and no missing data were recorded because questionnaires were checked for completeness on the spot.

The sample size was determined a priori using G*Power 3.1.9.2 software, which indicated a minimum of 324 participants at 95 percent power and an effect size of 0.402; the final cohort of 375 nurses comfortably exceeded this threshold. The demographic profile reflected the profession’s composition: 75.2 percent of respondents were female, 66.9 percent married, and 77.3 percent held bachelor’s degrees, with an average age of 34.60 years and an average of 11.43 years of professional experience. Participants were drawn from surgical and internal medicine wards, intensive care units, pediatrics, polyclinics, emergency services, and operating theatres. Because the score distributions deviated from normality, the team applied non-parametric statistics throughout, using the Kolmogorov-Smirnov test to confirm distribution shape, the Mann-Whitney U test for two-group comparisons, the Kruskal-Wallis H test for multiple groups with Bonferroni-corrected post hoc procedures, and Spearman correlation for continuous associations.

The headline result was a mean BNDS score of 54.62 out of a possible 63, a figure the authors characterize as relatively high and consistent with comparable studies using the same scale and with SPIKES-based assessments in other countries. Yet beneath that reassuring average lay striking disparities in preparation and role perception. Only 7.7 percent of nurses had ever received specific training in delivering bad news. Even more revealing, 85.6 percent believed that breaking bad news was not part of a nurse’s professional duty, and only four respondents—about one percent—considered it their responsibility. When asked to evaluate their own ability, 42.7 percent rated themselves as moderate, 20.3 percent as bad, and 14.4 percent as very bad, while fewer than a quarter described themselves as good or very good.

Statistical comparisons uncovered several significant moderators of perceived skill. Female nurses scored higher than male colleagues (mean 55.36 versus 52.37; p = 0.038, r = 0.124), a result consistent with a broader literature linking female healthcare professionals to stronger empathy and communication performance. Postgraduate-educated nurses outperformed both associate- and bachelor’s-degree holders (p = 0.002), echoing prior Turkish findings that advanced academic training confers an edge in therapeutic communication. Clinical setting mattered too: nurses in pediatric wards scored higher than those in internal medicine, and polyclinic nurses scored higher still (p = 0.009), possibly because outpatient staff interact with a larger and more diverse stream of patients and families, honing their empathic and problem-solving repertoire. Interestingly, whether a nurse had received special training made no significant difference to scale scores, a paradox the authors attribute to the fact that so few had such training and to the dominant influence of general vocational education.

Correlation analysis added a developmental dimension. Age showed a weak but significant positive association with BNDS scores (r = 0.178, p < 0.01), as did years of professional experience (r = 0.186, p < 0.01), suggesting that communication finesse in crisis conversations matures over a career, perhaps as nurses learn to manage their own emotions while navigating those of patients and relatives. Notably, the number of bad news deliveries per month—which averaged a mere 0.72—showed no relationship with skill scores, implying that frequency of exposure alone does not build competence in the absence of deliberate training and feedback. The findings align with earlier work reporting that nurses with two decades or more of experience communicate significantly better, though the literature is not unanimous; at least one study found communication skills declining with age, possibly reflecting occupational fatigue.

The discussion grapples with the most puzzling finding: nurses’ widespread denial that bad news delivery falls within their remit, even though the literature consistently frames it as a core nursing care activity. The authors suggest a self-reinforcing loop. Because nurses see the task as physicians’ responsibility, they do not seek specialized training; because they are untrained, they avoid the task and defer it upward, leaving patients’ emotional needs inadequately addressed. They also point to structural barriers nurses themselves cite, including limited time, fear of being unable to respond adequately, communication obstacles, violence from patients and relatives, and unsuitable environments. While established protocols—the SPIKES framework of Baile and Buckman, the ABCDE approach of Rabow and McPhee, and the algorithms of Girgis and Sanson-Fisher—offer guidance, the study notes these protocols are rarely implemented effectively in practice, and their success depends heavily on nurses’ education, experience, and institutional support.

Nurses who rated their own ability as good, moderate, or even bad scored higher on the scale than those who rated themselves very bad, while those reporting no negative experiences outscored those who were undecided—patterns suggesting that extreme self-doubt and ambiguity about one’s role both track with lower perceived competence. The authors interpret these effects cautiously, given the small effect sizes typical of the analysis (η²H values between 0.038 and 0.053, mostly below the conventional medium threshold) and the self-report nature of the data, which carries risks of social desirability bias, particularly with face-to-face interviews conducted by a researcher employed at the same hospital.

The study’s implications reach directly into nursing education and health policy. The authors call for structured communication training to be embedded in undergraduate curricula and in-service programs, for mentoring models that support nurses through emotionally difficult conversations, and for institutional policies that explicitly define nurses’ roles in breaking bad news and foster interdisciplinary collaboration. They further recommend that hospitals monitor bad news delivery as a quality and safety indicator. The limitations are candidly acknowledged: a single-center design that may not generalize, reliance on self-report without objective verification or patient outcome measures, a restricted set of variables, and a cross-sectional architecture that precludes causal inference. Even so, as hospitals worldwide confront workforce strain and rising expectations for patient-centered communication, the message from this Turkish cohort is clear: the professionals closest to the patient’s moment of distress are being asked to carry it largely unprepared, and closing that gap may be one of the most cost-effective investments a health system can make.

The study, published in Nature Human Behaviour, examined how people respond when healthcare providers describe symptoms as normal. Participants read realistic medical scenarios involving a broad range of conditions, including menopause symptoms, migraines, dental pain, seasonal allergies and elevated blood glucose levels. In some scenarios, providers used normalizing language; in others, they did not. The researchers then measured participants’ intentions to seek treatment and compared those responses with healthcare providers’ expectations.

The contrast was striking. Providers generally predicted that reassuring patients by describing their symptoms as normal would either increase their willingness to seek treatment or have no meaningful effect. Patients, however, frequently reacted in the opposite direction. Once a symptom was labeled normal, they were more likely to infer that medical intervention was unnecessary. In practical terms, a phrase intended to reduce anxiety appeared to lower treatment-seeking intentions.

The research grew partly from first author Seyi Lawal’s interest in communication surrounding menopause. Many people experiencing disruptive symptoms during menopause report feeling dismissed when told that their experiences are simply a normal part of aging. Lawal and her colleagues questioned whether such encounters reflected disagreement about the seriousness of symptoms—or a deeper difference in how patients and clinicians interpret everyday medical language.

Technically, the problem involves a shift between statistical and normative meanings. Clinicians often use “normal” descriptively, referring to symptoms that are common, predictable or well characterized within a population. Patients may interpret the word normatively, as a judgment that a condition is acceptable, benign or not deserving of treatment. The two meanings can coexist in a single conversation, but they lead to very different conclusions about what a patient should do next.

That distinction matters because the decision to seek care is shaped not only by symptoms themselves, but also by perceived medical necessity. If a patient interprets “normal” as “nothing is wrong,” the phrase can reduce the perceived benefits of diagnosis, monitoring or therapy. This may be especially important for conditions that are widespread but still painful, disabling or associated with long-term health risks. Commonness does not determine whether a symptom warrants attention, yet the study suggests that patients may unintentionally treat it as a signal that they should simply endure it.

The findings also speak to growing concerns about patients feeling ignored or invalidated in healthcare settings, experiences sometimes described as medical gaslighting. The researchers do not argue that clinicians are deliberately dismissive. Instead, they identify a communication mechanism through which well-intentioned reassurance can produce a dismissive experience. A doctor may believe they are normalizing a symptom to make a patient feel less frightened, while the patient may conclude that their concerns are not medically important.

The experiments tested ways to prevent that misunderstanding. One approach paired normalizing language with an explicit treatment recommendation, making clear that a symptom could be common while still deserving medical care. Another explained that “normal” referred to statistical frequency or clinical familiarity, rather than meaning that the symptom was desirable, harmless or something the patient should accept without help. Both strategies reduced the gap between what providers intended to communicate and what patients inferred.

Senior author On Amir said the findings do not mean doctors should stop reassuring patients. Rather, reassurance must be precise. A clinician might explain that a symptom is common and well understood, then immediately clarify whether evaluation, treatment, follow-up or lifestyle changes are recommended. For patients, the researchers offer a similarly direct message: hearing that a symptom is normal should not automatically be interpreted as advice to live with it. Asking what the clinician means—and whether treatment is appropriate—can prevent a single ambiguous word from shaping an important health decision.

The study, “Reassurance through normalization inadvertently suppresses treatment,” was funded in part by the T. Denny Sanford Institute for Empathy and Compassion. Its results suggest that a small change in clinical wording could have an outsized effect on patient behavior. In medicine, where decisions often depend on how risk and necessity are communicated, the difference between “common” and “not worth treating” may be only a few words—but those words can determine whether a patient seeks help.

Subject of Research: People

Article Title: Reassurance through normalization inadvertently suppresses treatment

Web References: https://www.nature.com/articles/s41562-026-02542-0

References: Nature Human Behaviour, DOI: 10.1038/s41562-026-02542-0

Keywords: health communication, doctor-patient relationship, treatment-seeking behavior, medical reassurance, symptom normalization, healthcare, verbal communication, medical gaslighting, patient decision-making, behavioral science