Indicators of Psychiatric Disorders in Different Oncology Specialties: A Prevalence Study

Objective. This study evaluated the prevalence of various indicators of psychiatric disorders in Brazilian outpatients with cancer and assessed possible associations with sociodemographic indicators. Materials and Methods. A total of 1,385 patients were evaluated using the following instruments: Patient Health Questionnaire-4 (PHQ-4), Generalized Anxiety Disorder (GAD-7), Fagerström Test for Nicotine Dependence (FTND), and Fast Alcohol Screening Test (FAST). Results. The sample was composed of both genders with a slight predominance of women (55.8%), subjects with incomplete/completed elementary school (59%), married (67.4%), with children (94%), not active from a labor viewpoint (61.6%), and following some type of religion (79.5%). The prevalence of anxiety for the total sample varied between 21.5 and 27.4%. The prevalence of depression was 21.1%, tobacco abuse/dependence was 40.2%, and alcohol was 20.3%. Women had significantly higher levels of anxiety and depression than men. Men had higher levels of substance abuse/dependence indicators than women. Conclusion. These results are consistent with the literature, which attests to the high prevalence of psychiatric disorder indicators in cancer patients, especially compared to the general population.


Introduction
The term "cancer" refers to a set of conditions that have the growth of cells that invade tissues and organs of the human body in common [1]. The causes of cancer are varied, but psychological and behavioral factors such as chronic stress, depression, and social isolation may contribute to the initiation and progression of certain cancers [2,3].
Conversely, cancer is a risk factor for the development of mental disorders. Symptoms of psychological and physical distress may emerge during cancer treatment in combination with disease systems [4][5][6].
Attention to the psychosocial aspects of the disease is equally important to cancer treatment [7], especially regarding psychopathologies because they significantly impact morbidity, low adherence to treatment, hospitalization duration, prognosis, quality of life, and patient survival [8][9][10][11].
Psychiatric disorders and symptoms may represent an adjustment reaction to the disease condition, and these symptoms may also be associated with a general medical condition, such as delirium [19]. Clinical complications and metabolic changes, such as hypercalcemia, anemia, vitamin B12 deficiency, and electrolyte imbalance, are also risk factors for mental disorders in cancer patients, especially depression [18].
Therefore, the impact of the cancer and the associated clinical conditions may promote the development of psychiatric disorders, especially mood disorders and psychotic symptoms [20,21].
Anxiety symptoms are commonly found in cancer patients. Pathological anxiety may be present in physical diseases, medication and drug use, alcohol withdrawal, central nervous system depressants, and the so-called anxiety disorders [22,23]. The prevalence of anxiety in cancer patients varies between 10 and 30% [24,25], and it is often associated with the diagnosis stage, procedure performance, and uncertainty about the future [20,21]. The abuse of psychoactive substances is directly associated with physical diseases, such as cirrhosis, cancer, and cardiovascular disease [26]. The consumption of alcohol and tobacco is a well-established and prominent etiological factor in patients with malignancy in the lungs and esophagus [27].
Various mental disorders are risk factors for the development of some cancers. Mental disorders may appear as comorbidities with the clinical condition, which may negatively impact disease diagnosis and treatment and emotional and financial costs [9][10][11].
However, oncology clinicians poorly identify mental disorders and their most common symptoms, despite the good responses to pharmacological and/or psychosocial interventions [28][29][30][31].
Knowledge of the prevalence of psychiatric disorders in this specific population may increase awareness and identification and mobilize resources for prevention and treatment [6,32]. Early screening for behavioral and/or psychological/psychiatric disorders in patients seen in health care settings positively affects health care quality and decreases suffering and institutional operational costs [33].
This study evaluated a statistically estimated sample of cancer outpatients at a hospital specializing in cancer according to the presence/absence of major mental symptoms/disorders using simple and quick screening instruments across different oncology specialties.

Subjects.
The sample comprised 1,385 subjects. Statistical calculations determined that a minimum sample size of 1,155 subjects was required with an error rate of 3%. The subjects were recruited from a single hospital specializing in cancer, and all subjects were receiving outpatient care (new or return cases) in different oncology specialties. The inclusion criteria were individuals of both genders over 18 years of age. The following exclusion criteria were adopted: severe cognitive impairment as qualitatively evaluated by the applicator and the absence of clinical conditions that would affect responses to the instruments.

Instruments.
The following self-assessment scales were used for data collection: (c) Fast Alcohol Screening Test (FAST) evaluates risky, harmful use and alcohol dependence syndrome. This version was translated and validated for Brazilian Portuguese [34]; (d) Fagerström Test for Nicotine Dependence (FTND) measures the degree of physical dependence on nicotine. This version was translated and validated into Brazilian Portuguese [35].
(e) Sociodemographic and Clinical Questionnaire assesses information about sociodemographic data (gender, schooling, civil status, professional situation, and religion) and treatments in mental health.

Data Collection and Analysis.
The subjects were selected randomly in reference to the scheduling system of the hospital according to the medical consultation dates. Subjects completed the questionnaires individually, and the evaluator remained available for assistance. The data were coded according to the ethical (Local Ethics Committee-Process no. 537/2011) and technical recommendations and entered into a database for analysis. The statistical package for social sciences (SPSS) version 19 was used. Descriptive (e.g., mean, standard deviation, and percentage) and parametric statistics (e.g., Student's -test, chi-square, and Pearson correlation test) were used for data analysis. A significance level of < 0.05 was adopted.

Results
The sample included adult individuals (mean = 50.3; DP = 13.9) of both genders with a slight predominance of women (55.8%). The sample included subjects with incomplete/completed elementary school (59%), married or in a stable relationship (67.4%), with children (86.7%), and not active from a labor viewpoint (61.3%). A total of 79.5% of the sample followed some type of religion, most often Catholicism (68%). Approximately 80% of the sample had no previous psychiatric/psychological history or psychiatric antecedents.
The prevalence of different indicators of psychiatric disorders in the total sample and according to different oncology specialties is shown in Table 1.
Anxiety's symptoms rates between 21.4% and 27.3% were observed for the total group, and Gynecology, Mastology, and Thorax specialties were prominent ( Table 1). The prevalence of depression indicators was 18.5% for the total group, and Gynecology, Orthopedics, and Thorax exhibited the highest rates.
The alcohol and tobacco abuse rates were 17.2% and 16.6%, respectively. The percentage prevalence of these abuse rates was higher in the Upper Digestive, Thorax, Head and Neck, and Urology specialties, reaching levels in excess of 30%.
Prevalence indicators for different sociodemographic characteristics of the sample are shown in Tables 2 and 3.
Women had higher anxiety indicators prevalence rates than men, and unmarried subjects exhibited higher anxiety rates than married subjects. This difference according to marital status was observed using the PHQ-4 A instrument, but it was not confirmed using the GAD-7 instrument.
Higher rates of symptoms of depression were associated with females, unmarried, and widowed subjects and subjects who were professionally inactive.
The rates of substance abuse (alcohol and tobacco) were more significant in men, married, and divorced individuals, subjects with less education, subjects who were not professionally active, and subjects who did not follow any religion ( Table 3).
The comorbidity rate in the total sample and among different oncology specialties is also noteworthy: 27.1% of the sample exhibited at least one comorbidity, with a prevalence of cooccurrence of anxiety and depression of 11.8%, and 6.6% of the sample had more than two comorbidities.
A total of 43.4% of patients in the Thorax/Lung specialty, 33.5% of the Upper Digestive specialty, and 29.8% of the Urology specialty exhibited indicators of more than one mental disorder.

Discussion
Cancer and psychiatric symptoms/disorders are strongly associated, and the present study evaluated this association in cancer outpatients undergoing treatment. Screening instruments evaluated symptoms of depression, anxiety, and substance abuse/chemical dependence.
The data showed depression indicators prevalence of 18.5%, which is greater than a previous Brazilian epidemiological study of 9.4% [31] but similar to a meta-analysis that reported a 16.3% depression rate in cancer patients at clinics and cancer hospitals [36]. These data corroborate previous reports of the prevalence of depression in this clinical group.
The prevalence of anxiety indicators varied between 21.4% and 27.3%, which are similar to primary care users (27.9%) as determined using observer-assessment interviews [37]. These data reinforce the important association between physical disease and anxiety, regardless of treatment context. Moreover, the rate of anxiety indicators in this study is high compared to a previous study that diagnosed 7.6% of cancer patients with an anxiety disorder using a diagnostic interview [38]. However, the choice of instrument may explain this difference, and the present study used screening instruments that measure anxiety symptoms rather than specific disorders.
The prevalence rate of alcohol abuse/dependence for the total sample was 20.3%. This level is higher than patients admitted to general medical specialties (9.8%) and hospitalized cancer patients (15.5%), who were all evaluated using a screening instrument [39].
The tobacco abuse rate for the total sample was 40.2%, which was similar to the prevalence of 44% in patients admitted to gastroenterology and pulmonology wards using screening scales and diagnostic interviews [40]. However, this rate is higher than the Brazilian population (24.1%) [37]. These findings suggest tobacco use as a possible risk factor, especially for oncological diseases of the respiratory system.
The Gynecology, Mastology, and Thorax specialties showed the highest rates of depression and anxiety indicators. One possible explanation for this finding is the exclusive presence of women in the first two specialties. Other evaluation studies of cancer patients reported that women subjects admitted to general hospitals or the general population exhibit high rates of depression and anxiety symptoms [31,[41][42][43][44][45]. It is possible too that, in Thorax specialty, the poorer prognosis can favor these rates [46].
One study evaluated 987 patients with lung cancer and found depressive symptoms in most subjects [47]. Another study evaluated patients with lung cancer using a selfadministered Hospital Anxiety and Depression Scale (HADS) and found that 37% had depression symptoms [48]. One possible explanation for this result is suggested by two other studies that showed an increased likelihood of depressive episodes in subjects during nicotine withdrawal, and an even greater likelihood for patients with a history of previous depressive episodes [49,50]. Therefore, patients with lung cancer may begin a process of reduction or remission of their smoking habit and develop withdrawal symptoms, which promotes the onset of depression and anxiety symptoms.
Men predominated among patients abusing alcohol and tobacco, which has also been demonstrated in previous evaluations in the general population in different countries [39,[51][52][53]. Therefore, these high rates and the behaviors associated with the male gender, who care less about health and engage in more risky behavior [54][55][56][57][58], support the use of these substances as a risk factor for disease development [27,[59][60][61][62][63]. The higher prevalence rates of alcohol and tobacco abuse in Thorax/Lung and Upper Digestive specialty patients and patients with tumors in the head and neck confirm this hypothesis.

Conclusions
The data presented are consistent with the previous literature and attest to the high prevalence of indicators of psychiatric disorders in cancer patients, especially compared to the prevalence in the general population. These results support the conclusion that individuals with cancer are more vulnerable to the onset and development of psychiatric disorders.
Conversely, the association between substance abuse and certain cancers indicates that substance abuse can be also a risk factor for disease development.
This study was conducted using a large and statistically estimated sample, but the psychiatric indicators were evaluated using screening instruments. These instruments were validated previously and show adequate sensitivity and specificity, but they only provide the prevalence of indicators but do not indicate disorders per se. The subjects' selfperception also influences these instruments, which may be overestimated or underestimated, and this fact should be considered in data interpretation as well as the fact that all the patients come from a single treatment center.
The high rates indicate the need to screen for psychiatric disorders in cancer patients, especially because of the damaging effects that psychiatric symptoms may exert on patient treatment. Active monitoring and early detection of symptoms may facilitate treatment adherence, decrease the duration of hospital stay, and assist in disease coping.
The study also highlights the need for greater investment in prevention campaigns that discourage the use of legal substances, such as alcohol and nicotine, because these substances are risk factors in cancer development.

Conflict of Interests
The authors declare that there is no conflict of interests regarding the publication of this paper.