Background
Although the incidence of gastric cancer has declined over the last few decades in many countries [
1‐
4], it is still one of the most common cancers in the world. In 2012, an estimated 951,600 newly diagnosed gastric cancer patients and 723,100 deaths occurred worldwide [
5]. Gastric cancer rates are generally about twice as high in men as they are in women [
6]. Most important, the incidence rates of gastric cancer vary widely across countries. Evidence has showed that incidence rates are highest in Eastern Asia, Central and Eastern Europe, and South America. In addition, more than half of newly diagnosed gastric cancer cases occurring in Eastern Asia [
5,
7]. For example, in 2012, there were 3795 Taiwanese being diagnosed as gastric cancer and the gastric cancer was the fifth-most frequently diagnosed cancer with a age-standardized incidence rate of 15.03 per 100,000 people [
8]. Moreover, gastric cancer has also a high mortality; for instance, in Taiwan, gastric cancer ranks as the sixth-highest cause of cancer-related deaths in 2012 with a standardized mortality rate of 6.4 per 100,000 people [
7].
Cancer is a risk factor for developing emotional disturbances, especially depression [
9]. Evidence has revealed that depressive disorders are associated with reduced quality of life and poorer outcomes in patients with cancer [
10]. Several studies found an increased risk of suicide among patients with cancer compared with the general population [
11‐
13]. Additionally, depression has also been associated with poor survival among gastric cancer patients in China [
14].
Depression is the most prevalent psychiatric disorder among the various types of cancer patients. Massieet et al. reviewed previous studies regarding the prevalence of depression in patients with cancer and found that the reported prevalence (major depressive disorder: 0–38%; depression spectrum syndromes: 0–58%) varied significantly because of different conceptualizations of depression, criteria used to define depression, methodological approaches to the measurement of depression, and study populations [
15]. Depression has been studied in patients with cancer using a range of assessment methods, such as self-report rating scales, brief screening instruments, and structured clinical interviews. The methods commonly used are the Hospital Anxiety and Depression Scale, Beck Depression Inventory, Hamilton Rating Scale for Depression questionnaire, and the Diagnostic and Statistical Manual of Mental Disorders criteria, published by the American Psychiatric Association. In general, the more narrowly the term is defined, the lower the reported prevalence of depression [
15]. Bergquist et al. found that the rate of depression among those with esophagus or gastroesophageal junction cancer ranged from 27 to 44% [
16]. Among these studies, diagnoses were usually not clinically defined and there have been few epidemiological studies specifically of depressive disorders among gastric cancer patients.
To reduce the evidence-to-practice gap and overcome the abovementioned shortage of previous studies and based on the fact that higher incidence and mortality rates of gastric cancer in Taiwan by annual government report and epidemiological studies, we designed a nationwide population-based retrospective cohort study, using a different definition for depressive disorders, to investigate the clinical pictures among the patients with gastric cancer and try to explore the differences on the incidence of subsequent depressive disorders compared with the matched cohort using the Taiwanese National Health Insurance Research Database (NHIRD). Also, independent predictive variables for subsequent depressive disorders following a gastric cancer diagnosis were also analyzed.
Discussion
The major finding of our study was that the risk of clinical depressive disorders requiring psychiatric intervention was higher among patients with gastric cancer compared to the matched cohort. In addition, female sex and hypertension were the risk factors for the development of subsequent depressive disorders following diagnoses of gastric cancer.
The current study design involved an unbiased participant selection process and used an age-, sex-, and comorbidity-matched cohort as the control group. Because participation in the NHI is mandatory, and nearly all Taiwanese residents can access healthcare with low copayments, referral bias is not a concern and follow-up compliance is high. Furthermore, to apply for a cancer catastrophic illness certificate, pathologic proof of malignancy is mandatory, and laboratory and imaging studies must be provided. Therefore, cancer diagnoses in this study were reliable. The strengths of this study were its large sample size, the long-term (12-year) follow-up period, and clinical depressive disorder diagnoses as defined based on the ICD-9-CM codes and antidepressant prescriptions.
Consistent with the results of previous studies, we observed that the risk of depressive disorders in gastric cancer patients was higher than it was in the control patients. The result may be explained by several possible factors. First, gastric cancer is often diagnosed at a late stage and its outcome is often poor with a 5-year survival rate under 10% [
21]; therefore, patients with gastric cancer may have a higher risk of experiencing emotional distress. Second, while the type of surgical procedure does not appear to impact quality of life (QOL), different types of gastric resection may affect eating behavior and physical and emotional functioning [
22]. For example, dumping syndrome is more common with a distal gastrectomy, and reflux is more common if the gastroesophageal junction is included in the specimen. These problems may lead to depressive disorders and other QOL problems. Third, studies have shown that patients with gastric cancer exhibit higher levels of mixed depression and anxiety symptoms compared to those with other types of cancer, raising the question of whether gastric cancer is associated with more biological mechanisms of depression, such as cytokine release [
23]. Consistent with this biological hypothesis, Koh et al. investigated the association between the brain-derived neurotrophic factor (BDNF) Val66Met polymorphism and stress coping response in patients diagnosed with gastric cancer and found that the BDNF Val66Met polymorphism may be involved in individual coping responses to cancer [
24].
Previous studies have reported that depressive symptoms were frequently noted in gastric cancer patients [
23,
25]. Although our results are consistent with such findings, the estimated incidence rate of depression among gastric cancer patients in these studies was approximately 20% [
25]; this rate is much higher than the rate calculated in our study. There are two possible reasons for the discrepancy between these findings. First, in previous studies, rating scales—such as the HAMD-24 questionnaire—rather than clinical diagnostic interviews, were used to identify depression. Patients who were recorded as depressed may not have fulfilled the diagnostic criteria for depressive disorders from a psychiatric perspective. In our study, we focused on the patients with depressive disorders who were prescribed antidepressants for at least 30 days by clinical oncologists or psychiatrists. Therefore, our study may reflect the real conditions of the gastric cancer patients under the care of clinical conditions. Second, cultural differences may impact the reporting of depressive symptoms in cancer patients [
26,
27]. Studies have shown that Asian patients focus more on their physical rather than their psychological symptoms [
28]. Moreover, in addition to the difficulty evaluating the etiology of physical symptoms in diagnosing depressive disorders, oncology specialists may also be uncertain regarding the effectiveness of treatment for depressive disorder, limiting referrals for psychiatric interventions. Therefore, gastric cancer patients with depressive disorders may have been underestimated in our study.
In our analysis of the risk factors associated with subsequent depressive disorders among patients with gastric cancer, female sex and hypertension were independent risk factors. The prevalence of hypertension was about 27.2% in 2000 in Asia [
29], and this has been increasing annually [
30]. From a psychosomatic perspective, hypertension could be concomitant with depression [
31]. Depression is closely associated with hypertension. Depression can promote the occurrence and development of hypertension [
32]. Meanwhile, hypertension is prone to aggravate depression, and the increased prevalence of depression has been described among patients with hypertension. Rabkin et al. found a three-fold higher frequency of major depressive disorder in patients treated for hypertension [
33]. These findings are in line with our study results, which indicated that hypertension may be a risk factor for subsequent depression among patients with gastric cancer. In addition, the female sex was at a greater risk for developing depressive disorders in the present study. This result is in line with previous studies in which the female sex was a risk factor for depression in patients with cancer [
34]. Moreover, the results of the analysis of the risk factors showed that treatment regimens were not associated with the increased risk of depressive disorders among patients with gastric cancer; these results seem consistent with earlier research on esophageal cancer patients [
16]. However, these results also highlight the complex life-life clinical situation and we should interpret the results carefully. For example, a gastric cancer patient who is eligible to receive curative surgery may have increased feelings of hope, possibly associated with the knowledge that the curative treatment implies a potential chance of a cure; however, the upcoming surgery and poor prognosis, despite undergoing curative treatment, may be reflected in heightened emotional distress. In addition, for those with gastric cancer that is diagnosed late and where surgical extirpation is unlikely, patients may focus on their physical illness needs and be reluctant or even hostile when asked about their psychological needs. These patients with later-stage gastric cancer were also expected to have a shorter life expectancy and, therefore, had fewer chances to receive psychiatric intervention.
Our study is one of the few nationwide studies to examine the association between gastric cancer and subsequent depressive disorders. Its strength is that it used a retrospective matched-cohort study design with a nationwide cohort of patients with gastric cancer and adequate controls for comorbidities. However, several limitations inherent to the use of medical care claims databases should be considered. First, depressive disorder causes are generally complex and vary depending on the patient. Many psychological and environmental factors can contribute to the development of these disorders. We acknowledge that several essential demographic variables—such as a family history of psychiatric disorders, stressful life situations, interpersonal relationships, and lifestyle—were unavailable in the medical care claims database. Second, the cancer stage could not be ascertained and, therefore, whether the severity of the gastric cancer influences the risk of developing depressive disorders warrants further study. Third, as mentioned previously, oncology specialists may view the depressive symptoms of gastric cancer patients as physical discomfort caused by gastric cancer and thus overlook these symptoms. The incidence of depressive disorders in our study may have been underestimated.