Abstract
Background
Psychiatric disorders are common in patients with cancer. The impact of both psychiatric disorders and psychiatric treatment on mortality in patients with cancer needs to be established.
Materials and Methods
Nationwide claims data were analyzed. To investigate the association between psychiatric disorders and mortality, 6,292 male and 4,455 female patients with cancer who did not have a record of psychiatric disorders before cancer onset were included. To examine the association between psychiatric treatment and mortality, 1,467 male and 1,364 female patients with cancer were included. Incident psychiatric disorder and receipt of psychiatric treatment within 30 days from the onset of a psychiatric disorder were the main independent variables. Dependent variables were all‐cause and cancer‐related mortality. Cox proportional hazards regression with time‐dependent covariates was used.
Results
The onset of psychiatric disorders was associated with a significantly increased risk of mortality in both male (all‐cause hazard ratio [HR]: 1.55; cancer‐related HR: 1.47) and female patients with cancer (all‐cause HR: 1.50; cancer‐related HR: 1.44) compared with patients with cancer without psychiatric disorders. Both male and female patients who received psychiatric treatment within 30 days of diagnosis of a psychiatric disorder had a lower risk of cancer‐related mortality (males, HR: 0.73; females, HR: 0.71) compared with patients with cancer with psychiatric disorders who did not receive psychiatric treatment.
Conclusion
Patients with cancer with newly diagnosed psychiatric disorders had a higher mortality rate. Among these, those who received psychiatric treatment showed lower rates of mortality. Thus, early detection and early treatment of psychiatric disorders in patients with cancer is needed.
Implications for Practice
The current study supplements the body of evidence supporting the association of psychiatric disorders onset and treatment with cancer outcomes. Patients with cancer showed an increased risk of both all‐cause and cancer‐related mortality upon psychiatric disorder onset. Among patients with newly diagnosed psychiatric disorders, those who received psychiatric treatment showed lower cancer‐related mortality. Thus, raising awareness of both the risk of psychiatric disorders and the positive effects of psychiatric treatment on cancer outcomes is necessary among patients with cancer, caregivers, and oncologists. Furthermore, it is necessary to adopt a multidisciplinary approach, encouraging patients with cancer to undergo a neuropsychological assessment of their mental health status and receive appropriate and timely psychological interventions.
Keywords: Psycho‐oncology, Cancer, Mortality, Psychiatric diagnosis, Psychotherapy
Short abstract
Cancer patients have higher rates of psychiatric disorders than the general population. This article reports on the association between psychiatric disorder onset and mortality among cancer patients, as well as the association between psychiatric treatment and mortality among cancer patients with psychiatric disorders.
Introduction
Cancer is a major issue for health authorities worldwide. Globally, it is estimated that in 2018, 18 million people were newly diagnosed with cancer, and 9.6 million died of it 1. In South Korea, cancer is a major health problem, as it has remained the leading cause of death for several decades. However, modern medical advances have increased the 5‐year survival rate for patients with cancer from 41.2% between 1993 and 1995 to 70.7% between 2011 and 2015 2.
As cancer survival rates increase, the mental health of patients with cancer is increasingly prominent. A cancer diagnosis is life‐threatening and can have major psychological effects. Patients’ awareness of their cancer diagnosis could lead to psychological stress 3, 4. Additionally, fear of death, grief about losses, and worries about friends and family could also contribute to psychological distress 5, 6. Furthermore, radiotherapeutic or chemotherapeutic interventions can also affect patients’ mental status. Such psychological distress could increase their vulnerability to mental health problems, which may affect their health behavior patterns and become diagnosable psychiatric disorders 7, 8. Thus, treating mental health is an important clinical oncology issue 9.
Patients with cancer have higher rates of psychiatric disorders than the general population 10, 11, including major depression, anxiety, adjustment disorders, delirium, and substance dependence disorders 9, 12. Approximately 30%–50% of patients with cancer suffer from psychiatric disorders 13. Particularly, depression is two to four times more common in patients with cancer than in the general population 14, 15.
In patients with cancer, psychiatric distress needs to be treated promptly, as delayed diagnosis or treatment may incur side effects, which could affect their health outcomes 16. Untreated psychological distress is associated with desire for death 17, disability 18, reduced quality of life 19, aggravated pain 20, diminished ability to plan end of life 21, and caregivers’ diminished psychosocial functioning 22 as well as prolonged hospitalization 19 and reduced cancer treatment adherence 17. Considering that psychiatric disorders among patients with cancer may be treatable 23, timely and appropriate approaches for diagnosis and treatment are needed to avoid negative consequences.
Despite numerous studies in recent decades focusing on depression and anxiety disorders in patients with cancer, few have examined the whole spectrum of psychiatric disorders or the effects of psychiatric treatment on clinical outcomes. Considering the important impact of psychiatric disorders on clinical outcomes, a study assessing psychiatric disorders from a broader perspective in relation to cancer mortality is warranted.
Thus, this study aimed to investigate the association between psychiatric disorder onset and mortality among patients with cancer. Moreover, we further investigated the association between psychiatric treatment and mortality among patients with cancer with psychiatric disorders.
Materials and Methods
Data and Study Population
This study used National Health Insurance claims data from 2002 to 2013. This data set included records from 1,025,340 individuals, representing approximately 2.2% of the Korean population 24. Patients’ claims data are recorded in four categories: insurance eligibility, medical institutions’ data, health examination data, and medical treatments (i.e., diagnosis code, medication, and treatment) 24.
Patients with gastric (C16), colorectal (C18–C19), lung (C34), and breast cancer (C50, only female)—the most frequent cancers in South Korea—were identified using International Classification of Disease 10th revision (ICD‐10) codes. Next, those who met the criteria for cancer, meaning that they visited a medical institution under an ICD‐10 C code at least three times within a year and/or were hospitalized in a medical institution under a C code, were selected 25. We excluded patients whose year of cancer diagnosis was 2002 (for wash‐out) or 2013 (to allow at least 1‐year follow‐up). We also excluded 2,245 men and 4,679 women who had an ICD F code (F00–F99), which indicates psychiatric disorder, prior to cancer onset, as we aimed to identify patients with new‐onset psychiatric disorders after cancer diagnosis. After excluding patients aged under 20 and those insured by medical aid, the sample comprised 6,292 men and 4,455 women. To investigate the association between psychiatric treatment and mortality, 1,515 men and 1,389 women diagnosed with psychiatric disorders after cancer onset were included. Among these, those who died within 30 days of diagnosis of psychiatric disorder were excluded from follow‐up of psychiatric treatment. Thus, the final sample was 1,467 men and 1,364 women (supplemental online Fig. 1). This study was approved by the Institutional Review Board of Yonsei University Health System (Y‐2018‐0065).
Variables
The dependent variables were all‐cause and cancer‐related mortality. Patients with death codes C16, C18–C20, C34, and C50 (for women) were considered to have died of cancer.
One of the main variables was postcancer psychiatric disorder onset. New‐onset psychiatric disorder was defined as a psychiatric disorder diagnosis in a patient who had no record of psychiatric disorder (ICD‐10 F codes) before cancer onset. For analysis, F codes were reclassified into six groups based on frequencies of psychiatric disorders among patients with cancer 26: (a) organic (including symptomatic) mental disorders (F00–F09); (b) mood disorders (F30–F39); (c) anxiety disorders (F40–F42); (d) neurotic, stress‐related, and somatoform disorders (F43–F45, F48, F50); (e) sleep disorders (F51, G47); and (f) other.
To investigate the effect of psychiatric treatment, patients with cancer who underwent psychotherapy or were prescribed psychiatric medication after psychiatric disorder onset were considered to have received psychiatric treatment; the time limit for receiving such psychiatric treatment was set as 30 days from psychiatric disorder diagnosis. If patients did not undergo psychotherapy or were not prescribed psychiatric medication within that time span, they were considered to have not received psychiatric treatment. The type of psychotherapy and medications were identified from treatment codes and generic codes (supplemental online Tables 1, 2).
To analyze the association between psychiatric diagnosis and mortality among patients with cancer, their age, income, insurance status, disability status, residential area, cancer site, Charlson comorbidity index, cancer treatment method, type, location, existence of a neuropsychiatry department in the diagnosing medical institution, and year of cancer onset were controlled for. Additionally, to analyze the association between psychiatric treatment and mortality, the type of psychiatric disorder and time to onset of psychiatric disorder (time period between cancer onset and psychiatric disorder onset) were controlled for (supplemental online Table 3).
Statistical Analysis
Chi‐squared tests were conducted to examine the distribution of general characteristics in the sample by calculating frequencies and percentages. Survival analyses were used for the main models. When analyzing cancer‐related mortality, all‐cause mortality was considered a competing risk. First, a Cox proportional hazards model with time‐dependent covariates was conducted to examine the association between psychiatric disorder onset and all‐cause/cancer‐related mortality among patients with cancer. Then, a Cox proportional hazards model was used to analyze the association between psychiatric treatment and all‐cause/cancer‐related mortality among patients with cancer with newly diagnosed psychiatric disorders.
Moreover, we conducted a sensitivity analysis by excluding patients newly diagnosed with organic (including symptomatic) mental disorders in order to examine whether the relationship between new‐onset psychiatric disorders and cancer‐related mortality would still hold. Treatment for psychiatric disorder was limited to within 30 days from the date of psychiatric disorder diagnosis, and patients who died within 30 days after psychiatric diagnosis were excluded from the landmark analysis. Sensitivity analyses were conducted by limiting treatment time to 3, 6, and 12 months from psychiatric diagnosis date. All statistical analyses were conducted separately by gender and using SAS software (version 9.4; SAS, Cary, NC).
Results
Table 1 shows the distribution of general characteristics in the study population. Of 6,292 male and 4,455 female patients with cancer, 1,515 males and 1,389 females experienced new‐onset psychiatric disorders. In men, 588 patients who developed new‐onset psychiatric disorders died (all‐cause death rate: 74.0 per 1,000 person‐years), with 447 dying of cancer (cancer death rate: 56.3 per 1,000 person‐years). In women, 318 patients who developed new‐onset psychiatric disorders died (all‐cause death rate: 39.2% per 1,000 person‐years), with 239 dying of cancer (cancer death rate, 29.5% per 1,000 person‐years; Table 1).
Table 1.
Hazard ratios for all‐cause and cancer‐related mortality with newly diagnosed psychiatric disorder after cancer onset
| Variables | Total | All‐cause mortality | Cancer‐related mortality | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| n | (%) | Person‐years | Death rate, per 1,000 person‐years | n | (%) | HR (95% CI)a | p value | Death rate, per 1,000 person‐years | n | (%) | HR (95% CI)a | p value | |
| Male | |||||||||||||
| Psychiatric disorder after cancer onset | |||||||||||||
| Onset | 1,515 | (24) | 7,944 | 74.0 | 588 | (39) | 1.55 (1.41–1.71) | <.0001 | 56.3 | 447 | (30) | 1.47 (1.32–1.63) | <.0001 |
| Nononset | 4,777 | (76) | 15,326 | 156.1 | 2,393 | (50) | 1 (Reference) | 133.6 | 2,048 | (43) | 1 (Reference) | ||
| Type of psychiatric disorder | |||||||||||||
| Organic, including symptomatic, mental disorders | 130 | (2) | 630 | 85.7 | 54 | (42) | 2.01 (1.53–2.65) | <.0001 | 55.5 | 35 | (27) | 1.65 (1.17–2.31) | .0039 |
| Mood disorders | 223 | (4) | 1,254 | 63.8 | 80 | (36) | 1.64 (1.31–2.06) | <.0001 | 45.4 | 57 | (26) | 1.48 (1.14–1.94) | .0039 |
| Anxiety disorders | 288 | (5) | 1,647 | 60.1 | 99 | (34) | 1.38 (1.12–1.69) | .0023 | 47.4 | 78 | (27) | 1.40 (1.11–1.76) | .0040 |
| Neurotic, stress‐related, and somatoform disorders | 313 | (5) | 1,825 | 58.1 | 106 | (34) | 1.37 (1.12–1.67) | .0021 | 42.2 | 77 | (25) | 1.26 (1.00–1.59) | .0522 |
| Sleep disorders | 468 | (7) | 2,141 | 97.6 | 209 | (45) | 1.53 (1.32–1.77) | <.0001 | 81.7 | 175 | (37) | 1.54 (1.31–1.80) | <.0001 |
| Others | 93 | (2) | 447 | 89.6 | 40 | (43) | 2.21 (1.61–3.02) | <.0001 | 56.0 | 25 | (27) | 1.69 (1.13–2.51) | .0102 |
| None | 4,777 | (76) | 15,326 | 156.1 | 2,393 | (50) | 1 (Reference) | 133.6 | 2,048 | (43) | 1 (Reference) | ||
| Male total | 6,292 | (100) | 23,270 | 128.1 | 2,981 | (47) | 107.2 | 2,495 | (40) | ||||
| Female | |||||||||||||
| Psychiatric disorder after cancer onset | |||||||||||||
| Onset | 1,389 | (31) | 8,112 | 39.2 | 318 | (23) | 1.50 (1.31–1.72) | <.0001 | 29.5 | 239 | (17) | 1.44 (1.23–1.67) | <.0001 |
| Nononset | 3,066 | (69) | 12,265 | 86.1 | 1,056 | (34) | 1 (Reference) | 73.1 | 897 | (29) | 1 (Reference) | ||
| Type of psychiatric disorder | |||||||||||||
| Organic, including symptomatic, mental disorders | 91 | (2) | 421 | 92.6 | 39 | (43) | 3.57 (2.57–4.96) | <.0001 | 64.1 | 27 | (30) | 3.04 (2.07–4.46) | <.0001 |
| Mood disorders | 263 | (6) | 1,513 | 41.0 | 62 | (24) | 1.41 (1.08–1.84) | .0109 | 34.4 | 52 | (20) | 1.42 (1.07–1.88) | .0042 |
| Anxiety disorders | 357 | (8) | 2,249 | 23.1 | 52 | (15) | 1.08 (0.82–1.44) | .5798 | 14.7 | 33 | (9) | 0.89 (0.64–1.24) | .4224 |
| Neurotic, stress‐related, and somatoform disorders | 248 | (6) | 1,640 | 26.2 | 43 | (17) | 1.20 (0.88–1.64) | .2560 | 20.7 | 34 | (14) | 1.23 (0.87–1.73) | .1790 |
| Sleep disorders | 369 | (8) | 1,974 | 50.7 | 100 | (27) | 1.68 (1.36–2.07) | <.0001 | 38.0 | 75 | (20) | 1.69 (1.35–2.11) | .0005 |
| Others | 61 | (1) | 315 | 70.0 | 22 | (36) | 1.60 (1.03–2.46) | .0353 | 57.2 | 18 | (30) | 1.68 (1.05–2.68) | .0315 |
| None | 3,066 | (69) | 12,265 | 86.1 | 1,056 | (34) | 1 (Reference) | 73.1 | 897 | (29) | 1 (Reference) | ||
| Female total | 4,455 | (100) | 20,377 | 67.4 | 1,374 | (31) | 55.8 | 1,136 | (26) | ||||
Adjusted for age group, income, insurance status, existence of disability, residential area, cancer site, Charlson comorbidity index, cancer treatment method, cancer diagnosed medical institution, existence of neuropsychiatry department, location of hospital, and cancer onset year.
Abbreviations: CI, confidence interval; HR, hazard ratio.
Regarding the type of psychiatric disorder, sleep disorder was the most frequent psychiatric diagnosis among both male (n = 468, 7.4%) and female patients (n = 369, 8.3%). The second most frequent psychiatric diagnosis was neurotic, stress‐related, and somatoform disorders among male patients (n = 313, 5.0%), and anxiety disorders among female patients (n = 357, 8.0%). The third most frequent psychiatric diagnosis was anxiety disorders among male patients (n = 288, 4.6%), and mood disorders among female patients (n = 263, 5.9%). The average time from cancer diagnosis to onset of psychiatric disorders was 877.7 days in men and 837.4 days in women (supplemental online Table 5).
Table 1 also shows the results of the Cox proportional hazards model with time‐dependent covariates for the association between psychiatric disorder onset and mortality. Psychiatric disorder onset was associated with a significantly increased risk of death in both men (all‐cause mortality hazard ratio [HR]: 1.55, p < .0001; cancer‐related mortality HR: 1.47, p < .0001) and women (all‐cause mortality HR: 1.50, p < .0001; cancer‐related mortality HR: 1.44, p < .0001) compared with patients with cancer without psychiatric disorders. This association was still observed after excluding patients with organic (including symptomatic) mental disorders in both men (all‐cause mortality HR: 1.51, p < .0001; cancer‐related mortality HR: 1.43, p < .0001) and women (all‐cause mortality HR: 1.49, p < .0001; cancer‐related mortality HR: 1.42, p < .0001; supplemental online Table 8).
Regarding the type of psychiatric disorder, the onset of organic (including symptomatic) mental disorders was associated with the highest risk of death in both men (all‐cause mortality HR: 2.01, 95% confidence interval [CI]: 1.53–2.65; cancer‐related mortality HR: 1.65, 95% CI: 1.17–2.31) and women (all‐cause mortality HR: 3.57, 95% CI: 2.57–4.96; cancer‐related mortality HR: 3.04, 95% CI: 2.07–4.46). In men, the onset of mood disorders was associated with the second highest risk of all‐cause mortality (HR: 1.64, 95% CI: 1.31–2.06) and third highest risk of cancer‐related mortality (HR: 1.48, 95% CI: 1.14–1.94). Onset of sleep disorders was associated with the third highest risk of all‐cause mortality (HR: 1.53, 95% CI: 1.32–1.77) and second highest risk of cancer‐related mortality (HR: 1.54, 95% CI: 1.31–1.80). In women, onset of sleep disorders was associated with the second highest risk of all‐cause mortality (HR: 1.68, 95% CI: 1.36–2.07) and cancer‐related mortality (HR: 1.69, 95% CI: 1.35–2.11).
When stratifying by cancer site, the highest risk of mortality for patients with cancer with new‐onset psychiatric disorders was observed for those with gastric cancer among both men (all‐cause mortality HR: 2.05, p < .0001; cancer‐related mortality HR: 1.89, p < .0001) and women (all‐cause mortality HR: 1.87, p < .0001; cancer‐related mortality HR: 1.81, p < .0001). In women, patients with breast cancer showed the second highest mortality upon psychiatric disorder onset (all‐cause mortality HR: 1.84, p = .0001; cancer‐related mortality HR: 1.77, p = .0012; Fig. 1).
Figure 1.
Hazard ratios for mortality associated with newly diagnosed psychiatric disorder by cancer site.Abbreviation: CI, confidence interval.
Table 2 shows the association between receiving psychiatric treatment within 30 days of psychiatric diagnosis and mortality. A total of 1,467 men were diagnosed with new‐onset psychiatric disorders. Of these, 1,097 received psychiatric treatment; among them, 397 patients died (all‐cause mortality), with 292 dying of cancer. Compared with those who did not receive treatment for new‐onset psychiatric disorders, those who received treatment showed lower HR (0.81, p = .0376). Regarding the type of psychiatric treatment, only pharmacotherapy showed a statistically significant association with mortality (all‐cause HR: 0.79, p = .0277; cancer‐related mortality HR: 0.71, p = .0039).
Table 2.
Hazard ratios for all‐cause and cancer‐related mortality with receiving psychiatric treatment within 30 days of the diagnosis of psychiatric disorder among the patients with newly diagnosed psychiatric disorder after cancer onset
| Variables | Total | All‐cause mortality | Cancer‐related mortality | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| n | (%) | Person‐years | Death rate, per 1,000 person‐years | n | (%) | HR (95% CI)a | p value | Death rate, per 1,000 person‐years | n | (%) | HR (95% CI)a | p value | |
| Male | |||||||||||||
| Receipt of psychiatric disorder treatment | |||||||||||||
| Yes | 1,097 | (75) | 3,446 | 115.2 | 397 | (36) | 0.81 (0.66–0.99) | .0376 | 84.7 | 292 | (27) | 0.73 (0.58–0.91) | .0051 |
| No | 370 | (25) | 853 | 179.4 | 153 | (41) | 1 (Reference) | 143.1 | 122 | (33) | 1 (Reference) | ||
| Type of psychiatric disorder treatment | |||||||||||||
| Both psychotherapy and pharmacotherapy | 85 | (6) | 294 | 126.0 | 37 | (44) | 0.85 (0.58–1.24) | .4018 | 85.1 | 25 | (29) | 0.73 (0.46–1.15) | .1776 |
| Psychotherapy only | 137 | (9) | 443 | 103.8 | 46 | (34) | 0.91 (0.64–1.29) | .5896 | 74.5 | 33 | (24) | 0.88 (0.58–1.34) | .5570 |
| Pharmacotherapy only | 875 | (60) | 2,709 | 115.9 | 314 | (36) | 0.79 (0.64–0.98) | .0277 | 86.4 | 234 | (27) | 0.71 (0.56–0.90) | .0039 |
| None | 370 | (25) | 853 | 179.4 | 153 | (41) | 1 (Reference) | 143.1 | 122 | (33) | 1 (Reference) | ||
| Male total | 1,467 | (100) | 4,299 | 128.0 | 550 | (38) | 96.3 | 414 | (28) | ||||
| Female | |||||||||||||
| Receipt of psychiatric disorder treatment | |||||||||||||
| Yes | 1,052 | (77) | 941 | 54.2 | 216 | (21) | 0.79 (0.59–1.05) | .1063 | 38.9 | 155 | (15) | 0.71 (0.51–0.98) | .0393 |
| No | 312 | (23) | 3,983 | 92.5 | 87 | (28) | 1 (Reference) | 75.5 | 71 | (23) | 1 (Reference) | ||
| Type of psychiatric disorder treatment | |||||||||||||
| Both psychotherapy and pharmacotherapy | 72 | (5) | 244 | 69.6 | 17 | (24) | 0.67 (0.38–1.17) | .1566 | 36.8 | 9 | (13) | 0.39 (0.19–0.82) | .0131 |
| Psychotherapy only | 166 | (12) | 715 | 46.2 | 33 | (20) | 0.67 (0.43–1.07) | .0947 | 35.0 | 25 | (15) | 0.56 (0.33–0.96) | .0355 |
| Pharmacotherapy only | 814 | (60) | 3,024 | 54.9 | 166 | (20) | 0.83 (0.61–1.12) | .2297 | 40.0 | 121 | (15) | 0.79 (0.56–1.12) | .1815 |
| None | 312 | (23) | 941 | 92.5 | 87 | (28) | 1 (Reference) | 75.5 | 71 | (23) | 1 (Reference) | ||
| Female total | 1,364 | (100) | 4,924 | 61.5 | 303 | (22) | 45.9 | 226 | (17) | ||||
Adjusted for age group, income, insurance status, existence of disability, residential area, cancer site, type of psychiatric disorder, Charlson comorbidity index, cancer treatment method, cancer diagnosed medical institution, existence of a neuropsychiatry department, location of hospital, time to onset of psychiatric disorder, and cancer onset year.
Abbreviations: CI, confidence interval; HR, hazard ratio.
In women, 1,364 patients were diagnosed with psychiatric disorders after cancer onset (Table 2). Of these, 1,052 patients received psychiatric treatment; among them, 216 died (all‐cause mortality), with 155 dying of cancer. Compared with those who did not receive treatment for new‐onset psychiatric disorders, those who received psychiatric treatment showed lower HR for cancer‐related mortality (0.71, p = .0393).
When stratifying by cancer sites, only men with colorectal cancer showed lower rates of cancer‐related mortality (HR: 0.42, 95% CI: 0.22–0.80) compared with patients with colorectal cancer who did not receive psychiatric treatment (Fig. 2). Figure 3 shows the results of the sensitivity analysis. When the time between psychiatric disorder onset and receipt of psychiatric treatment increased, the mortality risk also increased, although this association was not statistically significant.
Figure 2.
Hazard ratios for mortality associated with receiving psychiatric treatment within 30 days from the diagnosis of psychiatric disorder among patients with cancer with newly diagnosed psychiatric disorder by cancer site.Abbreviation: CI, confidence interval.
Figure 3.
Hazard ratios for mortality with psychiatric treatment receipt by time limit for receiving psychiatric treatment from the diagnosis of psychiatric disorder.
Discussion
The results of the present study showed that both male and female patients with cancer with new‐onset psychiatric disorders had a higher risk of mortality compared with patients with cancer without psychiatric disorders. Regarding the cancer site, both men and women with gastric cancer, and women with breast cancer, with new‐onset psychiatric disorders showed the highest mortality rates. Organic (including symptomatic) mental disorders, mood disorders, and sleep disorders were associated with a higher risk of mortality in both genders compared with those without new‐onset psychiatric disorders.
Prior studies have shown that depression, schizophrenia, substance‐abuse disorders, and other psychotic disorders increase the mortality rate of patients with cancer 27, 28, 29, 30. Furthermore, poor quality of cancer treatment could help explain their findings 31, 32, 33, 34, 35. For example, delirium produces symptoms of attention deficit and loss of awareness, which affects communication with family members and impedes participation in treatment‐related decisions, symptom assessment, and counseling; together, these factors in turn may aggravate the burden of symptom distress 36, 37. Moreover, patients with psychiatric disorders tend to lack social support 38, 39. Previous studies have demonstrated that patients with low quality or degree of social support and supportive relationships have lower survival rates 38, 39. Thus, the combination of these factors may adversely affect cancer treatment and render patients more vulnerable to negative cancer outcomes.
According to our results stratified by cancer site, both men and women with gastric cancer and women with breast cancer diagnosed with new‐onset psychiatric disorders showed the highest mortality rate compared with patients with cancer without psychiatric disorders. Usually, because of the comorbid dumping syndrome (in the case of gastrectomy) or gastrointestinal reflux, gastric cancer adversely affects eating behavior as well as emotional and physical functioning 40. On the other hand, breast cancer produces serious mental health problems among women, who may experience sexual dysfunction, concerns about infertility, menopausal syndrome, and emotional distress, including anxiety and depression, due to the negative effects of mastectomy on their body image 41, 42, 43. Furthermore, Watson et al. 30 found that patients with breast cancer who reported feelings of hopelessness or helplessness were more likely to die compared with those who did not experience such negative emotions.
In the current study, patients with cancer with newly diagnosed psychiatric disorders after cancer diagnosis showed lower rates of mortality when they received psychiatric treatment. Both men and women who received psychiatric treatment within 30 days of psychiatric diagnosis showed lower cancer‐related mortality compared with those with psychiatric disorders who did not receive treatment. Previous studies on this association have yielded conflicting results, with some indicating that psychotherapy was not associated with cancer outcomes, contrary to the present results 44, 45, 46, 47. Nevertheless, in one study, although group therapy had no association with survival, it improved quality of life 46. Numerous other studies showed positive associations between psychiatric treatment and cancer outcomes. For instance, it has been suggested that pharmacotherapy can potentially reverse the severity of cognitive symptoms in depressed patients with cancer 48. A recent study of patients with cancer found that depressive symptoms were associated with longer duration of hospitalization, but only in patients not taking antidepressant medication 49. Therefore, antidepressants could alleviate negative outcomes of patients with cancer with depression. Furthermore, some antidepressants, for example, mirtazapine, may also alleviate some chemotherapy‐related symptoms, such as nausea or cachexia‐anorexia 50.
In South Korea, stereotypes and prejudice may lead to insufficient awareness of the mental health needs of patients with cancer. Consequently, the optimal intervention timing may be overlooked 26. Moreover, patients with cancer are vulnerable to impaired mental health, and delay or absence of treatment for psychiatric disorders may worsen their impact on cancer outcomes. Studies on early psycho‐oncologic interventions, such as education about coping skills, demonstrated effective ways to control psychological distress, including depression and anxiety 51, 52; hence, early identification and treatment of psychological disorders is crucial.
The present results have several implications. First, awareness about the risk of psychiatric disorders should be promoted among patients with cancer, their caregivers, and oncologists. Because some psychiatric disorders, including depression and anxiety, could be regarded as natural consequences of cancer treatment, or may even be difficult to distinguish from chemotherapy side effects, many patients, caregivers, and oncologists tend to neglect these symptoms and leave them untreated 53. Thus, education about psychiatric symptoms and disorders needs to be provided at the cancer diagnosis stage or during treatment. Second, it is necessary to promote psychiatric consultation. Even if patients desire psychological support 54, it may be difficult to recognize their psychological symptoms, and only a small proportion of patients with cancer who experience psychological distress receive psychiatric interventions 55. Therefore, physicians should be aware of the types of patients who may be vulnerable to psychological distress and refer them for neuropsychiatric assessment/intervention in a timely manner. This requires a multidisciplinary approach; by connecting the oncology and neuropsychiatry departments and encouraging patients to undergo consultations, the mental health status of patients with cancer could be screened, and appropriate interventions could be offered. Third, considering that the average time from cancer diagnosis to onset of psychiatric disorders was 2–3 years, it is important to pay attention to patients with cancer who are in the survivorship period.
Limitations
The results of this study should be interpreted with caution because of several limitations. First, the limitations of administrative data due to potential inaccuracies have been previously discussed. For instance, ICD‐10 codes in cohort data may not always represent patients’ real disease status because their primary purpose is to facilitate patients’ health insurance claims 24. However, a previous study demonstrated a 70% correspondence between claims codes and medical record codes 56.
Second, because our study used administrative data, potential confounders that could affect mortality, such as health‐related behaviors (e.g., smoking, drinking, and physical activity) and presence of caregivers, could not be controlled for, as this information was not available.
Third, the exact stage of cancer could not be analyzed because the data set did not contain this information. Thus, it could be that patients with new‐onset psychiatric disorders were more ill and therefore had higher mortality. Similarly, patients with more advanced illness may have been less likely to receive psychiatric treatment and therefore also had increased mortality. However, to compensate for this limitation, cancer treatment methods (operation only, chemotherapy or radiotherapy before or after operation, and chemotherapy or radiotherapy only) were considered to approximately gauge cancer severity. However, this system may be imprecise because some chemotherapy drugs were not covered by the National Health Insurance during the study period.
Fourth, psychiatric disorder severity could not be included in this model because disorders were identified and classified using ICD‐10 codes. Furthermore, owing to the characteristics of the claim data, only patients who visited medical institutions with psychiatric symptoms and were diagnosed using a fixed set of psychiatric disorder classification codes were considered. Therefore, patients who exhibited psychiatric distress but did not visit medical institutions were not included. Thus, some study participants’ mental health status may have been misrepresented.
Fifth, psychotherapy was also identified and classified using treatment codes because medical records were not available. Therefore, it was impossible to determine the exact form of psychotherapy, therapy duration, who participated in therapy sessions, and other features.
Sixth, some psychiatric medications can target multiple symptoms; however, we could not separate them exactly.
Seventh, we could not classify “other” psychiatric disorders according to their type because of the small numbers of patients diagnosed in our sample.
Lastly, this study defined patients with newly diagnosed cancer and new‐onset psychiatric disorders using a set wash‐out period. However, this relied on an assumption because records preceding 2002 do not exist in an electronic format. Therefore, potential contamination of the inclusion criteria in the study population cannot be ruled out.
Conclusion
The current study supplements the body of evidence supporting the association of psychiatric disorder onset and treatment with cancer outcomes. This study found that patients with cancer with newly diagnosed psychiatric disorders showed higher mortality rates than those without a psychiatric diagnosis, regardless of gender. Furthermore, patients with cancer with newly diagnosed psychiatric disorders showed a lower risk of mortality when they received psychiatric treatment compared with those who did not receive it. Patients with cancer are vulnerable to psychiatric disorders, which can hamper the healing process and produce negative outcomes. Psychiatric treatment can be effective in alleviating the effects of psychiatric disorders among patients with cancer. Thus, raising awareness of both the risk of psychiatric disorders among patients with cancer and the positive effects of psychiatric treatment on cancer outcomes is necessary among patients with cancer, caregivers, and oncologists. Furthermore, through a multidisciplinary approach, encouraging patients to undergo neuropsychological assessment of their mental health and receive appropriate and timely psychological treatment is important to reduce the impact of psychiatric disorders on cancer outcomes. Further studies need to be conducted using information about the exact cancer stage to better understand the relationship between psychiatric disorders and cancer mortality.
Author Contributions
Conception/design: Eun‐Cheol Park
Provision of study material or patients: Sung‐In Jang
Collection and/or assembly of data: Tae Hyun Kim
Data analysis and interpretation: Sang Ah Lee, Chung Mo Nam
Manuscript writing: Sang Ah Lee, Young Hoon Kim
Final approval of manuscript: Sang Ah Lee, Chung Mo Nam, Young Hoon Kim, Tae Hyun Kim, Sung‐In Jang, Eun‐Cheol Park
Disclosures
The authors indicated no financial relationships.
Supporting information
See http://www.TheOncologist.com for supplemental material available online.
Supplemental Figures
Supplemental Tables
Disclosures of potential conflicts of interest may be found at the end of this article.
References
- 1. Bray F, Ferlay J, Soerjomataram I et al. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2018;68:394–424. [DOI] [PubMed] [Google Scholar]
- 2. Jung KW, Won YJ, Kong HJ et al. Cancer statistics in Korea: Incidence, mortality, survival, and prevalence in 2015. Cancer Res Treat 2018;50:303–316. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Derogatis LR, Morrow GR, Fetting J et al. The prevalence of psychiatric disorders among cancer patients. JAMA 1983;249:751–757. [DOI] [PubMed] [Google Scholar]
- 4. Grabsch B, Clarke DM, Love A et al. Psychological morbidity and quality of life in women with advanced breast cancer: A cross‐sectional survey. Palliat Support Care 2006;4:47–56. [DOI] [PubMed] [Google Scholar]
- 5. Besisik SK, Kocabey G, Caliskan Y. Major depression and psoriasis activation due to interferon‐α in a patient with chronic myeloid leukemia; “overlooked and/or misdiagnosed adverse reaction in malignant disease”. Am J Hematol 2003;74:224. [DOI] [PubMed] [Google Scholar]
- 6. Massie MJ. Prevalence of depression in patients with cancer. J Natl Cancer Inst Monogr 2004;2004:57–71. [DOI] [PubMed] [Google Scholar]
- 7. Cohen S, Kessler RC, Gordon LU. Strategies for measuring stress in studies of psychiatric and physical disorders In: Cohen S, Kessler RC, Gordon LU, eds. Measuring stress: A Guide for Health and Social Scientists. New York: Oxford University Press, 1995:3–26. [Google Scholar]
- 8. Kissane DW, Grabsch B, Love A et al. Psychiatric disorder in women with early stage and advanced breast cancer: A comparative analysis. Aust N Z J Psychiatry 2004;38:320–326. [DOI] [PubMed] [Google Scholar]
- 9. Akechi T, Nakano T, Okamura H et al. Psychiatric disorders in cancer patients: Descriptive analysis of 1721 psychiatric referrals at two Japanese cancer center hospitals. Jpn J Clin Oncol 2001;31:188–194. [DOI] [PubMed] [Google Scholar]
- 10. Kadan‐Lottick NS, Vanderwerker LC, Block SD et al. Psychiatric disorders and mental health service use in patients with advanced cancer: A report from the coping with cancer study. Cancer 2005;104:2872–2881. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Audrey M. Psychiatric disorder in a sample of the general population with and without chronic medical conditions. Am J Psychiatry 1988;145:976–998. [DOI] [PubMed] [Google Scholar]
- 12. Akechi T, Okamura H, Nishiwaki Y et al. Psychiatric disorders and associated and predictive factors in patients with unresectable nonsmall cell lung carcinoma: A longitudinal study. Cancer 2001;92:2609–2622. [DOI] [PubMed] [Google Scholar]
- 13. Mitchell AJ, Chan M, Bhatti H et al. Prevalence of depression, anxiety, and adjustment disorder in oncological, haematological, and palliative‐care settings: A meta‐analysis of 94 interview‐based studies. Lancet Oncol 2011;12:160–174. [DOI] [PubMed] [Google Scholar]
- 14. Walker J, Hansen CH, Martin P et al. Prevalence, associations, and adequacy of treatment of major depression in patients with cancer: A cross‐sectional analysis of routinely collected clinical data. Lancet Psychiatry 2014;1:343–350. [DOI] [PubMed] [Google Scholar]
- 15. Lutgendorf SK, Andersen BL. Biobehavioral approaches to cancer progression and survival: Mechanisms and interventions. Am Psychol 2015;70:186–197. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Rodin G. Effective treatment for depression in patients with cancer. Lancet 2014;384:1076–1078. [DOI] [PubMed] [Google Scholar]
- 17. Breitbart W, Rosenfeld B, Pessin H et al. Depression, hopelessness, and desire for hastened death in terminally ill patients with cancer. JAMA 2000;284:2907–2911. [DOI] [PubMed] [Google Scholar]
- 18. Portenoy R, Thaler H, Kornblith A et al. Symptom prevalence, characteristics and distress in a cancer population. Qual Life Res 1994;3:183–189. [DOI] [PubMed] [Google Scholar]
- 19. Ashbury FD, Madlensky L, Raich P et al. Antidepressant prescribing in community cancer care. Support Care Cancer 2003;11:278–285. [DOI] [PubMed] [Google Scholar]
- 20. Passik SD, Dugan W, McDonald MV et al. Oncologists' recognition of depression in their patients with cancer. J Clin Oncol 1998;16:1594–1600. [DOI] [PubMed] [Google Scholar]
- 21. Block SD. Assessing and managing depression in the terminally ill patient. Ann Intern Med 2000;132:209–218. [DOI] [PubMed] [Google Scholar]
- 22. Ballenger JC, Davidson JR, Lecrubier Y et al. Consensus statement on depression, anxiety, and oncology. J Clin Psychiatry 2001;62:64–67. [PubMed] [Google Scholar]
- 23. Montazeri A, Jarvandi S, Haghighat S et al. Anxiety and depression in breast cancer patients before and after participation in a cancer support group. Patient Educ Couns 2001;45:195–198. [DOI] [PubMed] [Google Scholar]
- 24. Lee J, Lee JS, Park SH et al. Cohort profile: The National Health Insurance Service–National Sample Cohort (NHIS‐NSC), South Korea. Int J Epidemiol 2017;46:e15. [DOI] [PubMed] [Google Scholar]
- 25. Hwangbo Y, Kang D, Kang M et al. Incidence of diabetes after cancer development: A Korean national cohort study. JAMA Oncol 2018;4:1099–1105. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. National Health Insurance Service Ilsan Hospital . A study on the risk of mental illness and the characteristics of psychiatric treatment in cancer patients in Korea: National Health Insurance Service Ilsan Hospital, 2014.
- 27. Mols F, Husson O, Roukema JA et al. Depressive symptoms are a risk factor for all‐cause mortality: Results from a prospective population‐based study among 3,080 cancer survivors from the profiles registry. J Cancer Surviv 2013;7:484–492. [DOI] [PubMed] [Google Scholar]
- 28. Pinquart M, Duberstein P. Depression and cancer mortality: A meta‐analysis. Psychol Med 2010;40:1797–1810. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Satin JR, Linden W, Phillips MJ. Depression as a predictor of disease progression and mortality in cancer patients: A meta‐analysis. Cancer 2009;115:5349–5361. [DOI] [PubMed] [Google Scholar]
- 30. Watson M, Haviland J, Greer S et al. Influence of psychological response on survival in breast cancer: A population‐based cohort study. Lancet 1999;354:1331–1336. [DOI] [PubMed] [Google Scholar]
- 31. Manderbacka K, Arffman M, Suvisaari J et al. Effect of stage, comorbidities and treatment on survival among cancer patients with or without mental illness. Br J Psychiatry 2017;211:304–309. [DOI] [PubMed] [Google Scholar]
- 32. Kisely S, Crowe E, Lawrence D. Cancer‐related mortality in people with mental illness. JAMA Psychiatry 2013;70:209–217. [DOI] [PubMed] [Google Scholar]
- 33. Kisely S, Sadek J, MacKenzie A et al. Excess cancer mortality in psychiatric patients. Can J Psychiatry 2008;53:753–761. [DOI] [PubMed] [Google Scholar]
- 34. Chang CK, Hayes RD, Broadbent MT et al. A cohort study on mental disorders, stage of cancer at diagnosis and subsequent survival. BMJ Open 2014;4:e004295. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Howard LM, Barley EA, Davies E et al. Cancer diagnosis in people with severe mental illness: Practical and ethical issues. Lancet Oncol 2010;11:797–804. [DOI] [PubMed] [Google Scholar]
- 36. Lawlor PG, Gagnon B, Mancini IL et al. Occurrence, causes, and outcome of delirium in patients with advanced cancer: A prospective study. Arch Intern Med 2000;160:786–794. [DOI] [PubMed] [Google Scholar]
- 37. Fainsinger RL, Tapper M, Bruera E. A perspective on the management of delirium in terminally ill patients on a palliative care unit. J Palliat Care 1993;9:4–8. [PubMed] [Google Scholar]
- 38. Goodwin JS, Hunt WC, Key CR et al. The effect of marital status on stage, treatment, and survival of cancer patients. JAMA 1987;258:3125–3130. [PubMed] [Google Scholar]
- 39. Kennedy S, Kiecolt‐Glaser JK, Glaser R. Immunological consequences of acute and chronic stressors: Mediating role of interpersonal relationships. Br J Med Psychol 1988;61:77–85. [DOI] [PubMed] [Google Scholar]
- 40. Karanicolas PJ, Graham D, Gönen M et al. Quality of life after gastrectomy for adenocarcinoma: A prospective cohort study. Ann Surg 2013;257:1039. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41. Ganz PA, Rowland JH, Desmond K et al. Life after breast cancer: Understanding women's health‐related quality of life and sexual functioning. J Clin Oncol 1998;16:501–514. [DOI] [PubMed] [Google Scholar]
- 42. Ganz PA. Psychological and social aspects of breast cancer. Oncology (Williston Park) 2008;22:642–646, 650. [PubMed] [Google Scholar]
- 43. Begovic‐Juhant A, Chmielewski A, Iwuagwu S et al. Impact of body image on depression and quality of life among women with breast cancer. J Psychosoc Oncol 2012;30:446–460. [DOI] [PubMed] [Google Scholar]
- 44. Goodwin PJ, Leszcz M, Ennis M et al. The effect of group psychosocial support on survival in metastatic breast cancer. N Engl J Med 2001;345:1719–1726. [DOI] [PubMed] [Google Scholar]
- 45. Musselman DL, Somerset WI, Guo Y et al. A double‐blind, multicenter, parallel‐group study of paroxetine, desipramine, or placebo in breast cancer patients (stages I, II, III, and IV) with major depression. J Clin Psychiatry 2006;67:288–296. [DOI] [PubMed] [Google Scholar]
- 46. Kissane DW, Grabsch B, Clarke DM et al. Supportive‐expressive group therapy for women with metastatic breast cancer: Survival and psychosocial outcome from a randomized controlled trial. Psychooncology 2007;16:277–286. [DOI] [PubMed] [Google Scholar]
- 47. Cunningham A, Edmonds C, Jenkins G et al. A randomized controlled trial of the effects of group psychological therapy on survival in women with metastatic breast cancer. Psychooncology 1998;7:508–517. [DOI] [PubMed] [Google Scholar]
- 48. Twillman RK, Manetto C. Concurrent psychotherapy and pharmacotherapy in the treatment of depression and anxiety in cancer patients. Psychooncology 1998;7:285–290. [DOI] [PubMed] [Google Scholar]
- 49. Wong RL, El‐Jawahri A, D'Arpino SM et al. Use of antidepressant medications moderates the relationship between depressive symptoms and hospital length of stay in patients with advanced cancer. The Oncologist 2019;24:117–124. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50. Riechelmann RP, Burman D, Tannock IF et al. Phase II trial of mirtazapine for cancer‐related cachexia and anorexia. Am J Hosp Palliat Care 2010;27:106–110. [DOI] [PubMed] [Google Scholar]
- 51. Faller H, Schuler M, Richard M et al. Effects of psycho‐oncologic interventions on emotional distress and quality of life in adult patients with cancer: Systematic review and meta‐analysis. J Clin Oncol 2013;31:782–793. [DOI] [PubMed] [Google Scholar]
- 52. Schneider S, Moyer A, Knapp‐Oliver S et al. Pre‐intervention distress moderates the efficacy of psychosocial treatment for cancer patients: A meta‐analysis. J Behav Med 2010;33:1–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53. Söllner W, DeVries A, Steixner E et al. How successful are oncologists in identifying patient distress, perceived social support, and need for psychosocial counselling? Br J Cancer 2001;84:179. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 54. Merckaert I, Libert Y, Messin S et al. Cancer patients' desire for psychological support: Prevalence and implications for screening patients' psychological needs. Psychooncology 2010;19:141–149. [DOI] [PubMed] [Google Scholar]
- 55. Pascoe S, Edelman S, Kidman A. Prevalence of psychological distress and use of support services by cancer patients at Sydney hospitals. Aust N Z J Psychiatry 2000;34:785–791. [DOI] [PubMed] [Google Scholar]
- 56. Park B, Suh S, Sung J et al. Improvement plan for validity of health insurance disease code and establishment of data application plan: Seoul National University College of Medicine, Health Insurance Review & Assessment Service, 2003. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
See http://www.TheOncologist.com for supplemental material available online.
Supplemental Figures
Supplemental Tables