Associations Between Demographic, Clinical, and Symptom Characteristics and Stress in Oncology Patients Receiving Chemotherapy

INTRODUCTION

Stress is a common human experience that activates the hypothalamic-pituitary-adrenal axis so an individual can adapt to the experience. However, unrelieved acute and chronic stress result in poorer health outcomes.¹ A cancer diagnosis and its treatments are regarded as highly stressful events. Chronic stress can adversely impact patients’ prognosis and their quality of life.^{2, 3}

Throughout the oncology literature,^{4, 5} findings regarding oncology patients’ experiences of stress are highly variable depending on when stress is evaluated during the course of the cancer trajectory. However, it is recognized that multiple sources of stress (e.g., global stress, cancer-specific stress) can have a negative impact on patients’ physical, psychological, and social well-being, as well as their overall quality of life. For example, the initial cancer diagnosis, as well as the stress associated with subsequent treatments (e.g., scheduling appointments, waiting for results of diagnostic tests, fear of unrelieved symptoms and adverse effects) and fears of recurrence, are considered cancer-specific or disease-specific stress. This type of stress may provoke symptoms associated with post-traumatic stress disorder (PTSD).⁶

This cancer-specific stress is distinct from global stress that arises from the ordinary pressures of daily life.⁶ For example, global stressors may include: the time needed to care for children or older adults, challenges at work, and difficulties with routine home maintenance activities. These global stressors contribute to decrements in oncology patients’ ability to function and in their overall quality of life. Most patients undergoing cancer treatment experience and manage both types of stress.^{7, 8}

While both types of unrelieved stress can have a negative impact on all aspects of oncology patients’ lives,^9–11 few studies have evaluated for common and unique demographic, clinical, and symptom characteristics associated with elevated levels of global and cancer-specific stress in the same sample of patients. While younger age and unmarried/unpartnered status were associated with both types of stress,^{7, 12} female gender, being White, and having a lower socioeconomic status were associated with higher cancer-related distress.⁷ In terms of clinical characteristics, most studies have focused on patients with breast cancer. In a systematic review of 42 studies of breast cancer survivors,¹³ a lower functional status, a higher number of comorbidities, receipt of chemotherapy, and disease recurrence were associated with higher levels of psychological distress. In contrast, among newly diagnosed women with breast cancer,¹⁴ while clinical characteristics were not associated with cancer-specific distress, negative illness perceptions were a significant risk factor. Regarding the relationships between stress and symptoms, a growing body of evidence suggests that higher levels of stress are associated with increases in pain,¹⁵ fatigue,¹⁶ sleep disturbance,¹⁷ chemotherapy-induced neurotoxicities,¹⁸ depression,¹⁹ and cognitive impairment²⁰ in patients with a variety of chronic conditions. Additional research is needed to determine which demographic, clinical, and symptom characteristics are associated with global stress and/or cancer-specific stress in oncology patients.

Therefore, the purpose of this study, in a sample of oncology patients with heterogenous types of cancers (n=941), was to use the same set of demographic, clinical, and symptom characteristics to evaluate their relative contribution to the severity of global (i.e., measures using the Perceived Stress Scale (PSS)²¹) and cancer-specific (i.e., measured using the Impact of Event Scale-Revised (IES-R)²²) stress. We hypothesized that common and distinct characteristics would be associated with global and cancer-specific stress.

METHODS

RESULTS

DISCUSSION

This study is the first to evaluate for associations between a consistent and comprehensive set of demographic, clinical, and symptom characteristics and multiple dimensions of stress in a large sample of oncology patients with heterogenous types of cancers. Congruent with our a priori hypothesis, several common and distinct characteristics were associated with measures of global and cancer-specific stress. Moreover, consistent with our a priori hypothesis and prior research,⁸ our findings suggest that global and cancer-specific stress are distinct types of stress.

Our patients’ mean scores on the PSS (18.4 ±8.1) and the IES-R (18.6 ±12.9) are comparable to previous findings among oncology patients undergoing chemotherapy. For example, in a study that examined the relationship between stress and quality of life subsequent to a cancer diagnosis and its treatment among breast cancer patients,⁶ the mean PSS score was 18.1 (±6.9) with a range of 0–36. In another study of the associations between global and cancer-specific stress and symptoms among cancer survivors,³⁸ the mean PSS score was 17.3 (±8.9) and approximately 8.2% of survivors reported an IES-R score of >33. While the mean IES-R total score of our sample fell below the cutoff for probable PTSD, 29.4% of our patients had scores above the clinically meaningful cutoff score of ≥24.0 and 13.9% had scores of ≥33.0, suggestive of probable PTSD. Taken together, our findings suggest that a significant proportion of our patients (i.e., 43.3%) were experiencing clinically meaningful levels of cancer-specific stress. Clinicians need to be aware that occurrence of PTSD symptoms warrants a referral for additional evaluation and psychological services if needed.

Except for the IES-R avoidance subscale (i.e., 15.6%), the remaining regression analyses explained between 41.6% (i.e., IES-R total score) and 54.5% (i.e., PSS score) of the total variance in the stress measures. As noted in one study,³⁹ compared to the IES-R intrusion and hyperarousal subscales, the items evaluated on the avoidance subscale do not fit the Diagnostic and Statistical Manual of Mental Disorders (DSM) criteria for PTSD. While the avoidance subscale focuses on active avoidance strategies, the DSM clusters the symptoms of avoidance and emotional numbing together. Therefore, given the small amount of explained variance for the avoidance subscale, future research needs to include additional characteristics (e.g., coping strategies, personality, mental adjustment to cancer) to determine their relative contribution to the severity of this type of stress.

A higher level of depressive symptoms was the only characteristic that was associated with higher levels of stress across all our measures and made the largest contribution to the percentage of explained variance. Depression explained between 13.3% and 20.1% of the total variance in the stress scales. It should be noted that while the mean CES-D score of the sample was below the clinically meaningful cutoff, 31% of the patients had CES-D scores of ≥16. While the other characteristics in the various models were statistically significant, they only explained less than 1% or not more than 3% of the total variance in the various stress scales. This finding is congruent with the growing body of evidence that describes linkages between stress and depression.⁴⁰ As noted in one review,⁴⁰ prolonged stress can produce maladaptive neuroendocrine changes that can lead to the development of depression. Findings from other studies suggest that given the variety of biological processes that link depressive symptoms with sustained stress responses (e.g., dysregulation of the hypothalamic pituitary axis and the immune system; excessive activation of the sympathetic nervous system and excitatory neurotransmitters; loss of neuroplasticity),¹¹ multiple feedback loops may occur that render it difficult to discern directionality. Taken together with our finding of a significant percentage of our patients with PTSD symptoms, clinicians need to assess for sources of significant stress, as well as depression during and following chemotherapy administration.

Relative to the associations between various demographic and clinical characteristics, a number of the most common symptoms associated with cancer and its treatments uniquely explained a higher percentage of the total variance in the various stress scales. Overall, our findings concur with previous studies that found positive associations between pain,¹⁵ fatigue,¹⁶ sleep disturbance,¹⁷ cognitive impairment ²⁰ and stress in patients with a variety of chronic conditions. In terms of global stress, our findings are congruent with previous studies that suggested that global stress from environmental factors cumulatively influences decrements in mental health and cognitive function.⁴¹ Additional research is warranted to examine the potential impact of stress reduction strategies to improve mental health and cognition. However, clinicians can recommend a variety of stress reduction techniques (e.g., relaxation, meditation) to oncology patients and assess the effects of these interventions on stress, as well as overall symptom burden.

Our study found that, pain, fatigue, and sleep disturbance were associated with cancer-specific stress. Our findings are consistent with increasing evidence that demonstrates that these three symptoms as well as depressive symptoms often occur as a cluster ⁴² and are associated with cancer-specific stress. For example, in a study of patients with renal cell carcinoma,⁴³ symptoms of depression and cancer-specific post-traumatic stress were independently associated with fatigue and sleep disturbance. Moreover, the associations with these two symptoms were even stronger when depression and cancer-related post-traumatic stress co-occurred. In addition, our findings are consistent with studies that reported on the co-occurrence of pain and PTSD symptoms.^{44, 45} Considered together, the positive associations between these frequently co-occurring symptoms and global and cancer-specific stress highlight the need for continued research to investigate their common and distinct mechanisms and directionality of the relationships.

In our study, the demographic characteristics associated with higher levels of global stress and cancer-specific stress were consistent with prior research. Similar to previous research that found an association between lower socioeconomic status and higher cancer-related distress,¹³ we found an association between lower levels of education and higher levels of cancer-specific stress (i.e., higher IES-R Avoidance scores). These two demographic characteristics may be proxy measures for various social determinants of health that are known to be associated with higher levels of stress and poorer outcomes in oncology patients.⁴⁶

In addition, as noted previously,^{12, 47} younger age was associated with higher levels of global and cancer-specific stress. Extending prior findings, our analyses of the IES-R subscales suggest that younger age was associated with higher levels of hyperarousal but not with avoidance or intrusion. Living alone was associated only with higher levels of hyperarousal. Further research is needed to determine whether the associations between younger age and living alone and higher IES-R hyperarousal scores may reflect differences between younger and older adults in illness perceptions, coping, and social support that could lessen typical symptoms of hyperarousal (e.g., anger, reckless behavior, exaggerated startle, hypervigilance).

While prior studies found that not being married/partnered was associated with higher levels of both global and disease specific stress,¹² our findings suggest that this characteristic was associated with only cancer-specific stress (i.e., higher IES-R total, avoidance, and intrusion scores). This finding may be explained by the emotional and practical support (e.g., accompaniment to appointments, treatment decisions, household chores) provided by spouses or partners that may mitigate some of the stress related to a cancer diagnosis and its treatment.

Of note, having child care responsibilities was the only characteristic associated with higher levels of global stress. These responsibilities increase the pressures of daily life (e.g., increased financial burden, role overload, relationship conflict) and contribute to overall life stress.⁴⁸ Future research needs to evaluate the efficacy of interventions that support caregiving responsibilities in an effort to reduce stress during cancer treatment.

Limited research exists on the associations between clinical characteristics and stress. While in one study, no associations were found,¹⁴ in other studies, decrements in functional status,^{13, 49} increased number of comorbidities,^{13, 50} receipt of chemotherapy,¹³ and disease recurrence ¹³ were associated with higher levels of stress. Among the clinical characteristics evaluated in this study, a higher number of metastatic sites including lymph node involvement was the only characteristic associated with both global and cancer-specific stress. This finding is most likely explained by the fact that the presence of metastatic disease is associated with changes in treatment ⁵¹ and/or a poorer prognosis.⁵² It is not readily apparent why a lower BMI and a lower MAX2 score (i.e., less toxic chemotherapy regimen) were associated with higher avoidance scores. These findings warrant confirmation in future studies.