Increases in Stress and Adverse Childhood Experiences Are Associated With the Co-occurrence of Anxiety and Depression in Oncology Patients

Patients and Settings

For this longitudinal study,¹¹ eligible patients were ≥18 years of age; had a diagnosis of breast, gastrointestinal, gynecological, or lung cancer; had received chemotherapy within the preceding four weeks; were scheduled to receive at least two additional cycles of chemotherapy; were able to read, write, and understand English; and gave written informed consent. Patients were recruited from two Comprehensive Cancer Centers, one Veteran’s Affairs hospital, and four community-based oncology programs. A total of 2234 patients were approached and 1343 consented to participate (60.1% response rate). The major reason for refusal was being overwhelmed with their cancer treatment.

Instruments

Study Procedures

Study was approved by the Institutional Review Board at the University of California, San Francisco and each of the study sites. Eligible patients were approached by a member of the research team in the infusion unit to discuss participation in the study during their first or second cycle of chemotherapy. Written informed consent was obtained from all patients. Patients completed the anxiety and depression measures six times over two cycles of chemotherapy (i.e., prior to chemotherapy administration (Assessments 1 and 4); one week following the administration of chemotherapy (Assessments 2 and 5); two weeks after the administration of chemotherapy (Assessments 3 and 6)). All of the other measures were completed at enrollment (i.e., prior to the second or third cycle of chemotherapy). Medical records were reviewed for disease and treatment information.

Data Analysis

LPA was used to identify subgroups of patients with distinct joint state anxiety AND depression profiles. Using Mplus version 8.4,²⁹ this LPA was done with the combined set of variables over time (i.e., using the STAI-S AND CES-D scores obtained during the six assessments in a single LPA). This approach provides a profile description of these two symptoms with parallel profiles over time.

Model fit was evaluated to identify the solution that best characterized the observed latent class structure with the Bayesian Information Criterion, Vuong-Lo-Mendell-Rubin likelihood ratio test, entropy, and latent class percentages that were large enough to be reliable.³⁰ Missing data were accommodated for with the use of the Expectation-Maximization algorithm.³¹ Additional analyses were done using SPSS version 28 (IBM Corporation, Armonk, NY). Bonferroni corrected p-value of <0.017 was considered statistically significant for the three possible pairwise contrasts (i.e., .05/3).

Latent Class Solution

The rational for the selection of the 3-class solution is described in Table 1. The classes were named based on clinically meaningful cutpoints for state anxiety and depression. As illustrated in Figure 1, of the 1328 patients, 57.5% were in the Low Anxiety and Low Depression class (Both Low); 33.7% in the Moderate Anxiety and Moderate Depression class (Both Moderate); and 8.8% in the High Anxiety and High Depression class (Both High).

Demographic and Clinical Characteristics

Compared to Both Low class, the other two classes were younger; more likely to self-report being of Hispanic, Mixed race or other ethnicity; more likely to live alone, and more likely to be unemployed (Table 2). Compared to Both Low class, Both High class was more likely to report a past or current history of smoking and self-reported having lung disease, ulcer or stomach disease, or anemia or blood disease.

Differences among the three classes in being unmarried or partnered and having a lower annual income followed the same pattern (Both Low < Both Moderate < Both High). In terms of clinical characteristics, differences among the three classes in KPS scores (Both Low > Both Moderate > Both High) and number of comorbidities, SCQ scores, and the occurrence of depression and back pain (Both Low < Both Moderate < Both High) followed similar patterns.

Stress and Resilience

Differences among the classes in PSS, IES-R total and intrusion, avoidance, and hyperarousal subscales, and LSC-R total, affected sum, and PTSD sum scores followed the same pattern (i.e., Both Low < Both Moderate < Both High; Table 3). Differences among the three classes in CDRS scores were as follows: Both Low > Both Moderate > Both High.

Occurrence of Life Stressors

Compared to Both Low class, the other two classes reported higher occurrence rates for physical abuse at <16 years, physical abuse at ≥16 years, being forced to touch at < 16 years, and being forced to touch at ≥16 years (Table 4). Compared to the other two classes, Both High class reported higher occurrence rates for family violence in childhood and physical neglect. Compared to Both Low class, Both High class reported higher occurrence rates for sexual harassment, forced sex ≥16 years, having a family member in jail, having a serious physical or mental illness other than cancer, and caring for someone with a severe physical or mental handicap. Differences among the three classes in the occurrence rates for emotional abuse and serious money problems followed the same pattern (i.e., Both Low < Both Moderate < Both High).

Effects of Life Stressors

Compared to Both Low class, the other two classes reported higher effect scores for been in a serious accident, having serious money problems, being separated from a child, caring for someone with a severe physical or mental handicap, experiencing the sudden death of someone close, and experiencing the death of someone closed that was not sudden (Table 5). Compared to Both Low class, Both High class reported higher effect scores for family violence in childhood, physical abuse at ≥16 years, forced to touch at <16 years, seeing in a serious accident, being separated/divorced, and seeing a robbery/mugging. Differences among the three classes in the effect of emotional abuse was as follows: Both Low < Both Moderate < Both High.

Coping

In terms of the engagement coping strategies, compared to Both Low class, the other two classes reported lower scores for active coping, positive reframing, and acceptance (Table 6). Compared to Both Low class, Both High class reported lower scores for use of emotional support. In terms of disengagement strategies, compared to Both Low class, other two classes reported higher scores for the use of self-distraction and substance use. For the use of denial, behavioral disengagement and venting, the scores followed the same pattern (Both Low < Both Moderate < Both High).

Global and Cancer-Specific Stress

All of the stress measures exhibited a dose response effect (i.e., as the anxiety AND depression profiles worsened, all of the stress scores increased; Table 3). In terms of global stress, no clinically meaningful cutoff score is available for the PSS. However, scores for our Both High class were comparable to those reported by patients who were assessed within three to six months following breast cancer surgery.³⁴ In terms of cancer-specific stress, patients in Both Moderate and Both High classes had IES-R total scores suggestive of PTSD symptomatology and probable PTSD, respectively. These high scores are of concern in patients receiving chemotherapy because in a sample of men with a past history of PTSD,³⁵ IES-R scores of ≥24 were associated with immunosuppression.

As noted in a meta-analysis,³⁶ prevalence rates for PTSD in oncology patients vary from 7.3% to 13.8%. This range is lower than the 20.6% to 66.1% found for patients in the Both Moderate and Both High classes. Plausible explanation for these higher rates of PTSD include that our patients were undergoing chemotherapy and were experiencing a high symptom burden of between 8 to 23 unrelieved symptoms (out of a total of 38).³⁷

Implications of the strong associations among anxiety, depression, and stress warrant careful consideration. In a review of neuroimaging studies that examined the relationships between “negative affect” (i.e., distress) and changes in metabolism or structure of brain regions,³⁸ cancer patients showed changes in function and/or structure in key brain regions (e.g., prefrontal cortex, amygdala, thalamus, hippocampus, hypothalamus, insula, basal ganglia) that are associated with greater anxiety, depression, PTSD, and stress. The authors noted that knowledge of these inter-relationships provides insights into the effects of various psychological factors on peripheral stress-related pathways that are known to contribute to cancer progression and poorer long-term health outcomes. This information supports the need for the prescription of stress management interventions for a substantial number of oncology patients.

Stressful Life Events (SLEs)

In addition to the high levels of global and cancer-specific stress, particularly in Both High class, the number and effects of SLEs suggest that these patients were experiencing extremely high levels of stress. While the mean number of stressors reported by the Both High class was 8.0, 25% of these patients experienced between 11 and 23 stressors. Of note, occurrence rates for experiencing family violence in childhood, physical neglect, sexual harassment, having a serious mental or physical illness, caring for someone with a severe physical or mental handicap, or having a family member who was incarcerated were significantly higher in Both High class compared to Low class.

An examination of the specific SLEs, suggest trends in higher occurrence rates for specific stressors in Both Moderate and Both High classes. One specific trend is the higher occurrence rates for adverse childhood experiences (ACE’s). These stressors are described as traumatic experiences that overwhelm an individual’s ability to cope at the time of its occurrence. They are associated with negative physical and psychological health outcomes, as well as with difficulty developing positive coping behaviors.^{39, 40}

It is well established that exposure to ACE’s increases the risk for major depression and anxiety in adulthood.⁴¹ The experience of SLEs results in increases in inflammation and allostatic load that are associated with the occurrence of a variety of chronic conditions including cancer.⁴² As noted in one meta-analysis,⁴³ compared to patients with no ACEs, individuals with 2 or 3 kinds of ACEs or at least 4 ACEs had a 1.35 to 2.17 increased risk of cancer, respectively. Among the ACEs that were evaluated in this review and found in our sample, physical abuse, sexual abuse, exposure to intimate partner violence, and financial difficulties were associated with an increased risk for cancer in adulthood. Equally important, in one population-based study,⁴⁴ a synergistic effect was found between exposure to ACEs and being female and the occurrence of a major depression episode and an anxiety disorder.

Resilience

Resilience is defined as an individual’s ability to maintain or restore relatively stable physical and psychological functioning when confronted with an adverse experience or SLEs.⁴⁵ Similar to the stress measures, resilience scores demonstrated a dose response effect. Patients in Both Moderate and Both High classes had CDRS scores that were below the normative score for the US population. As noted in one review,⁴⁶ cancer patients with lower levels of resilience had higher levels of psychological distress and poor adjustment to their cancer. In addition, in a review of resilience in survivors of critical illness,⁴⁷ estimates of the occurrence rates for lower levels of resilience ranged from 28% to 67% which is consistent with the 42.5% found in our study. In addition, lower levels of resilience were associated with higher levels of anxiety, depression, and post-traumatic stress.

Coping

Both Moderate and Both High classes reported higher utilization of disengagement and lower utilization of engagement type coping strategies. These findings are consistent with a study of women with breast cancer that reported that higher use of denial and disengagement strategies were associated with higher levels of distress.⁴⁸ Compared to Both Low class, Both High class had lower utilization of emotional support as a coping strategy. This result is consistent with a study of women with breast cancer that found that lower levels of social support were associated with higher levels of anxiety and depression.⁴⁹ In addition, given that a higher percentage of patients in Both High class were less likely to be married/partnered and more likely to be living alone, as well as less likely to use emotional support suggests that loneliness may contribute to a higher psychological symptom burden. This hypothesis is supported by studies that found positive associations between loneliness and anxiety⁵⁰ and depression.⁵⁰

Demographic and Clinical Characteristics

Consistent with previous reports, younger age,⁵¹ identifying as female,³² and being single³³ were associated with membership in Both Moderate and/or Both High classes. In terms of younger age, this result is consistent with a study that evaluated for mixed anxiety/depression in cancer patients and found significant reductions in the co-occurrence of both symptoms with each 10-year increase in age.¹ As noted in one study,³² younger individuals may have less clearly established social networks; have less experience navigating the healthcare system; and have more challenges with transportation, finances, and child care that contribute to higher levels of depression and anxiety.

Patients in Both Moderate and Both High classes were less likely to be married/partnered and more likely to live alone. Our findings are consistent with the extant literature that demonstrates higher levels of anxiety and depression, as single symptoms, are associated with higher levels of loneliness and social isolation as well as decreases in social support.⁵²

Being unemployed and having a lower annual household income were additional risk factors for being in the two worst profiles. Of note, 23.6% of the patients in these two classes reported an annual income of <$30,000. In addition, these patients reported higher occurrence rates for “having serious money problems” on the LSC-R. Our results are consistent with a study of ovarian cancer patients that found that lower income was an independent and the most relevant risk factor for persistent depression following chemotherapy.³³ In addition, in one registry study, co-occurrence of socioeconomic deprivation and multimorbidity was associated with higher rates of depression and anxiety.⁵³

Lower functional status and higher comorbidity burden were found in Both Moderate and Both High classes. These findings are consistent with several reviews that noted positive associations between depression and anxiety and multimorbidity.^54–56 Taken together, these demographic and clinical characteristics are often linked with other social determinants of health. While these risk factors are generally not modifiable, these patients warrant referrals to social services for assistance.

Limitations

The majority of the sample was white, well-educated, and reported a moderate to high income. Therefore, future studies need to enroll a more diverse sample. Given that the mean age of the sample was in the mid 50’s, the co-occurrence of anxiety and depression in younger patients warrants consideration. Finally, the primary reason for lack of participation was being overwhelmed with cancer treatment, which suggests that our findings may underestimate the co-occurrence of these two symptoms in patients receiving chemotherapy.

Implications for Practice

Findings from this study suggest that in addition to assessing for anxiety and depression, clinicians need to assess for SLEs and ACEs on a routine basis. Identification of high risk patients will allow for the prescription of both pharmacologic and non-pharmacologic interventions for one or both symptoms, as well as for stress. Prescription of timely interventions is a priority because previous research found that while mortality rates were similar in patients with lung cancer who never reported or had treatment for depression, patients without treatment had higher mortality rates.⁵⁷ In addition, clinicians can provide education on the use of engagement coping strategies to reduce symptom burden and stress. High risk patients may benefit from cognitive behavioral therapy and/or mindfulness-based interventions to establish and build their coping skills. Study findings add support for the need to integrate mental health and social services across the continuum of cancer care.²