Researchers led by G. Elise Doppenberg-Smit of Amsterdam University Medical Center report in the journal Supportive Care in Cancer that two previously overlooked features of emotional life, instability and variability, reliably distinguish cancer patients whose emotions have turned maladaptive from those whose emotions remain adaptive. The team tracked 132 patients with solid tumors for 14 consecutive days, pinging their mobile phones five times a day to capture momentary ratings of eight negative emotional symptoms, including feeling irritated, nervous, anxious, down, and worried. It is the first study to examine emotion dynamics in patients with cancer, and its findings could reshape how clinicians decide who needs professional mental health care.

The technical logic behind the study draws on a growing field known as emotion dynamics research. Rather than treating emotions as static quantities, researchers model them as time series, using statistical measures that describe how feelings fluctuate across dozens of moments. Emotional inertia, typically quantified by autocorrelation, captures the tendency of an emotion to carry over from one moment to the next, so that sadness or anxiety lingers and perpetuates rather than fading. Emotional variability, measured by the standard deviation, describes the range of a person’s emotions, or how far they deviate from their own average. Emotional instability, calculated through the root mean square successive difference, or RMSSD, quantifies the magnitude of rapid shifts, so that a person with high instability experiences sudden swings between emotional states.

These three parameters have been studied extensively in patients with depression, anxiety disorders, and borderline personality disorder, where a landmark meta-analysis of more than 11,000 subjects linked them to psychological well-being and psychopathology. But they had never been examined in oncology. The Dutch team recruited patients from four academic and two nonacademic hospitals between August 2021 and April 2024, enrolling people who had started systemic treatment three to seven months earlier. To sort patients into adaptive and maladaptive groups, the researchers used two independent reference standards: a full-structured psychiatric diagnostic interview assessing current depressive and anxiety disorders, and a self-report questionnaire capturing whether patients received or wished to receive a referral for professional mental health care.

The results were striking. Under the psychiatric diagnostic standard, patients with maladaptive emotions showed significantly higher emotional variability across all eight symptoms, with Cohen’s d effect sizes ranging from 0.49 to 0.84, indicating moderate to large effects. Emotional instability followed the same pattern, with effect sizes between 0.48 and 0.79. Under the subjective need standard, the pattern held for six of the eight symptoms, with moderate effects between 0.50 and 0.75 for variability and 0.55 to 0.80 for instability. In plain terms, patients whose emotions had become dysfunctional did not simply feel worse on average; their feelings swung more widely and shifted more abruptly from moment to moment, mirroring the extreme and unstable emotional profiles documented in patients with psychiatric conditions.

Emotional intensity, the traditional measure used in oncology screening tools such as the Distress Thermometer, also separated the groups. Patients with maladaptive emotions rated all symptoms as more intense under the psychiatric standard, with effect sizes between 0.58 and 1.06. Under the subjective need standard, the differences were less pronounced, and only half of the symptoms reached statistical significance. This asymmetry carries an important clinical message: the sheer strength of an emotion is especially indicative of a diagnosable disorder, and less so of a patient’s felt need for psychological support.

The hypothesis about emotional inertia, by contrast, largely failed. Patients with adaptive and maladaptive emotions differed in inertia for only two symptoms, feeling irritated and feeling down, under the psychiatric standard, and showed no significant differences at all under the subjective need standard. Part of the explanation lies in an unexpected data pattern: in nearly 18 percent of patient-symptom combinations, the standard deviation across all 70 assessments was zero, meaning patients consistently answered that a symptom was not present at all during the two-week window. These flat, absent emotions were far more common among patients with adaptive emotions, but their prevalence reduced the statistical power available for the inertia analyses. The researchers note that previous studies on inertia in depression have produced mixed results, leaving its status as an indicator of maladaptive emotions uncertain.

Robustness checks strengthened the core findings. When the team repeated the analyses separately for patients with metastatic and non-metastatic disease, the results resembled those of the full sample, though wider confidence intervals from the reduced sample sizes sometimes produced non-significant differences. A second assessment wave, conducted three months after the first, reproduced the pattern in a subset of 110 patients: variability and instability remained higher in the maladaptive group, while inertia showed inconsistent differences. The study’s use of Ecological Momentary Assessment, capturing emotions in real time across 70 equidistant moments rather than at a single clinic visit, is a key strength, since retrospective questionnaires are vulnerable to recall bias and miss the fluctuations the new measures are designed to detect.

The clinical implications could be far-reaching. Current screening methods that rely on emotional intensity face a well-documented problem: many patients who score above validated cut-offs do not want a referral or do not use psychological care, while additional indicators are needed to distinguish those who genuinely need support. The new findings suggest that instability and variability, alongside intensity, could serve as complementary signals. Before these dynamics can enter routine practice, the researchers caution, threshold values and diagnostic accuracy must be established for each indicator individually and in combination, and their integration into broader symptom monitoring systems must be tested against clinicians’ own impressions. The team also acknowledges limitations, including a smaller maladaptive group than anticipated and wide confidence intervals that underscore the need for replication in fresh samples.

Still, the study opens a provocative new chapter in psycho-oncology. If the way emotions move, rather than merely how strongly they are felt, marks the transition from adaptive coping to dysfunction, then the mathematics of emotional fluctuation may soon become as important to cancer care as the biology of the tumor itself. For the minority of patients whose feelings have turned against them, a phone that notices the swings could be the first step toward getting help.

Subject of Research: Emotion dynamics as indicators of maladaptive emotions in patients with cancer

Article Title: Emotion dynamics as indicators of maladaptive emotions in patients with cancer

Article References: Doppenberg-Smit, G. E., Lamers, F., Braamse, A. M. J., Pruijt, H., Kuip, E. J. M., Werter, I. M., van de Wetering, R. A. W., Verheul, H. M. W., Beekman, A. T. F., Dekker, J., Hoogendoorn, A. W., & van Linde, M. E. (2026). Emotion dynamics as indicators of maladaptive emotions in patients with cancer. Supportive Care in Cancer, 34(11), Article 1103. https://doi.org/10.1007/s00520-026-11320-7

Image Credits: AI Generated

DOI: 10.1007/s00520-026-11320-7

Keywords: emotion dynamics, cancer, psycho-oncology, ecological momentary assessment, mental health, emotional instability, emotional variability, Supportive Care in Cancer, Emotion, dynamics, indicators, maladaptive