Abstract
Objective
We investigate the prevalence of five affective temperaments (depressive, cyclothymic, hyperthymic, irritable, and anxious) in a large sample of cancer patients and associations of temperament with cancer site as well as the impact of temperament on overall survival of cancer patients.
Methods
Data for this prospective cohort study was collected in the outpatient clinic of a large cancer center. We used the Temperament Evaluation in Memphis, Pisa and San Diego – Münster Version (TEMPS-M) and recorded patient data. The sample consisted of 2531 patients with seven different cancer/disease-sites. Kruskal-Wallis tests and pairwise Wilcoxon rank sum test were applied to compare temperament scales across disease groups. For analyzing survival time, we used a Cox regression model and log-rank tests.
Results
The five affective temperaments were similarly distributed across all disease groups. We found higher levels of depressive, cyclothymic, and anxious temperament in women and higher levels of hyperthymic and irritable temperament in men. Temperament was mostly not predictive of survival, with only two significant results in the regression models. Here, cyclothymic temperament was predictive of mortality in the full sample and hyperthymic temperament was predictive of the pancreatic cancer subsample.
Conclusions
Our study provides evidence to debunk the myth of a cancer-specific temperament. Neither did we find a temperament profile that was different from studies with general population samples, nor were there any disease-specific profiles differentiating various types of cancer.
We investigate the prevalence of five affective temperaments (depressive, cyclothymic, hyperthymic, irritable, and anxious) in a large sample of cancer patients and associations of temperament with cancer site as well as the impact of temperament on overall survival of cancer patients.
Methods
Data for this prospective cohort study was collected in the outpatient clinic of a large cancer center. We used the Temperament Evaluation in Memphis, Pisa and San Diego – Münster Version (TEMPS-M) and recorded patient data. The sample consisted of 2531 patients with seven different cancer/disease-sites. Kruskal-Wallis tests and pairwise Wilcoxon rank sum test were applied to compare temperament scales across disease groups. For analyzing survival time, we used a Cox regression model and log-rank tests.
Results
The five affective temperaments were similarly distributed across all disease groups. We found higher levels of depressive, cyclothymic, and anxious temperament in women and higher levels of hyperthymic and irritable temperament in men. Temperament was mostly not predictive of survival, with only two significant results in the regression models. Here, cyclothymic temperament was predictive of mortality in the full sample and hyperthymic temperament was predictive of the pancreatic cancer subsample.
Conclusions
Our study provides evidence to debunk the myth of a cancer-specific temperament. Neither did we find a temperament profile that was different from studies with general population samples, nor were there any disease-specific profiles differentiating various types of cancer.
| Original language | English |
|---|---|
| Article number | 112015 |
| Journal | Journal of Psychosomatic Research |
| Volume | 189 |
| Early online date | 12 Dec 2024 |
| DOIs | |
| Publication status | E-pub ahead of print - 12 Dec 2024 |
Keywords
- Affective temperament
- Oncology
- Personality
- Psycho-oncology
- Survival analysis
ASJC Scopus subject areas
- Clinical Psychology
- Psychiatry and Mental health
