Prevalence of depressive disorders among head-and-neck cancer patients: A hospital-based, cross-sectional study.

BACKGROUND: Head-and-neck cancers (HNCs) are associated with significant psychosocial challenges at all stages of illness, which influence the course and outcome of cancer. We aimed to assess the prevalence of depressive disorders among patients with HNC and its sociodemographic and clinical determinants.
MATERIALS AND METHODS: It was a cross-sectional study conducted in the department of oncology of a medical college hospital from South India. A total of 100 adult patients with HNC were recruited over a period of 1 year, after obtaining ethical clearance. Mini International Neuropsychiatric Interview was used to assess for depressive disorders in these patients.
RESULTS: We found that 49% of the patients were suffering from major depressive disorder (MDD), 13% of the patients had MDD with melancholic features, and 10% had dysthymia. Functional impairment and surgical treatments were associated with the diagnosis of depressive disorders.
CONCLUSION: Depressive disorders are highly prevalent in HNC and emphasize the need for tailored psycho-oncological and psychotherapeutic interventions.

INTRODUCTION

Head-and-neck cancers (HNCs) account for 4% of all malignancies.[1] In India, HNCs are the most common cancer among men and the third most common among women.[2] Majority of the patients remain undiagnosed and do not receive adequate psychiatric care, which might affect the treatment compliance and cause overall poor quality of life. Many depressed patients adhere poorly to treatment recommendations, leading to reduced chances of survival.[3] Among all malignancies, patients with HNC have the highest rates of major depressive disorder (MDD). The incidence of depression in patients with HNC ranges from 15% to 50%. It has been hypothesized that a major reason is the physical location affected by HNC. The most basic aspects of one's life, including the ability to speak, eat, breathe, and appear in public, may all be affected by either the cancer or its treatment. Facial disfigurement cannot be concealed, thus creating significant impairment in social interactions and emotional expressions, which can further result in social withdrawal and avoidance of potentially helpful support systems.[4]

Research on psychiatric morbidities associated with HNC is limited in India and needs attention. In this study, we set out to explore depressive disorders in patients with HNC. We also want to explore the sociodemographic, clinical, and treatment factors, which are associated with depressive disorders in these patients.

MATERIALS AND METHODS

The study was conducted in the department of oncology of a medical college hospital from South India after obtaining ethical clearance. It was an observational, cross-sectional study. A total of 100 adult inpatients with a diagnosis of HNC who gave written informed consent were recruited using randomized table number method. Patients who were acutely ill or with impaired cognitive function were excluded from the study. Specially designed questionnaire was used by the principal investigator to record sociodemographic and clinical variables. Mini International Neuropsychiatric Interview version 5 for the Diagnostic and Statistical Manual of Mental Disorders IV and ICD 10 was used to assess the presence of depressive disorders.

Statistical analysis was done using Statistical Package for the Social Sciences version 23 (IBM Corporation., New York, USA). Descriptive statistics were used for reporting sociodemographic and clinical variables. Categorical variables were displayed in terms of percentages and proportions. Continuous variables were expressed as means and standard deviation (SD). Chi-square test and Fisher's exact test were used to explore the association between sociodemographic and clinical variables with psychiatric diagnoses.

RESULTS

Sociodemographic profile

The mean age of the patients included in this study was 50 years (SD ± 11.8 years). Majority of the patient sample was male (64%). Most of the patients were married (90%) and were living with their spouses at the time of interview, whereas only 10% were single, widow, or widowers. At the time of diagnosis of HNC, 63% were not working, whereas only 37% were working actively [Appendix Table 1].

Appendix Table 1

Sociodemographic variables (n=100)

Variable	Categories	Number of patients (%)
Sex	Male	64
	Female	36
Education	Less than SSC	79
	More than or equal to SSC	21
Marital status	Married	90
	Single/widow/widower	10
Occupation	Unemployed	29
	Self-employed	38
	Unskilled	26
	Semiskilled	3
	Skilled	4
	Student	0
Per capita income (rupees)	<5000	25
	5000-10,000	60
	>10,000	15
Socioeconomic status (modified BG Prasad classification)	Upper class	0
	Upper middle class	0
	Middle class	36
	Lower middle class	36
	Lower class	28
Health insurance	Yes	60
	No	40

Clinical profile

Clinical variables of the patient sample are presented in Table 1. Substance use was common among the sample studied, with 60% of the patients using tobacco, 40% chewing betel nut, and 28% using alcohol. Majority of the patients (80%) were given the diagnosis of cancer recently (<6 months).

Table 1

Clinical variables (n=100)

Variables	Categories	Number of patients (%)
Weeks after diagnosis	8 weeks	40
	9 weeks-24 weeks	40
	25 weeks and above	20
Cancer staging	Early (Stages I and II)	56
	Advanced (Stages III, IVA, IVB)	44
Treatment received	Surgery	25
	CCRT/CC/RT*	50
	Surgery + CCRT/CC/RT†	25
Diagnosis revealed to patient	Yes	78
	No	22
Activity status	No limitation	26
	Consider limitation	74

*Concurrent chemotherapy + radiotherapy/chemotherapy/radiotherapy; †Surgery + concurrent chemotherapy + radiotherapy/chemotherapy/radiotherapy

Majority of the patients (74%) experienced considerable limitations (effects on functions such as taste, swallowing, speech, breathing, and eating), and the rest of the patients (26%) did not report of any limitations in their activity.

There was no significant association between different sociodemographic variables and diagnosis of major depressive episode (MDE), MDE with melancholic features, and dysthymia using Pearson's Chi-square test/Fisher's exact test [Appendix Tables 2–4].

Appendix Table 2

Association between sociodemographic variables and diagnosis of major depressive episode

Sociodemographic variable	MDE		χ2/Fisher’s exact test	P
	Present	Absent
Sex
Male	30	34	0.321	0.571†
Female	19	17
Age
≤40	10	11	0.442	0.802†
41-59	28	26
≥60	11	14
Education
Less than SSC	39	40	0.20	0.887†
More than or equal to SSC	10	11
Marital
Married	45	45	-	0.741‡
Single/widow/widower	4	6
Occupation
Unemployed	14	15	0.009	0.926†
Employed	35	36
Work
Yes	14	23	2.928	0.087†
No	35	28
Religion
Hindu	34	37	1.179	0.658†
Muslim	15	13
Christian	0	1
Socioeconomic status
Middle class	13	23	4.136	0.126†
Lower middle class	19	17
Lower class	17	11

†Chi-square test, ‡Fisher’s exact test. MDE – Major depressive episode

Appendix Table 4

Association between sociodemographic variables and diagnosis of dysthymia

Sociodemographic variable	Dysthymia		Fisher’s exact test	P
	Present	Absent
Sex
Male	8	56	-	0.323
Female	2	34
Age
≤40	0	21	3.433	0.173
41-59	6	48
≥60	4	21
Education
Less than SSC	10	69	-	0.115
More than or equal to SSC	0	21
Marital
Married	9	81	-	1.000
Single/widow/widower	1	9
Occupation
Unemployed	2	27	-	0.719
Employed	8	63
Work
Yes	3	34	-	0.741
No	7	56
Religion
Hindu	7	64	0.910	1.000
Muslim	3	25
Christian	0	1
Socioeconomic status
Middle class	7	29	6.589	0.034
Lower middle class	3	33
Lower class	0	28

Table 2 shows the association between clinical variables and diagnosis of MDE using Pearson's Chi-square test. There was statistically significant association between the activity status of the patient and the diagnosis of MDE using Pearson's Chi-square test (P < 0.001). Patients with HNC who had considerable limitation of their activity status were more likely to be diagnosed with MDE.

Table 2

Association between clinical variables and diagnosis of major depressive episode

Clinical data	MDE		χ2 test	P
	Present	Absent
Cancer staging
Early	24	32	1.922	0.166
Advanced	25	19
Treatment received
Surgery	10	15	1.321	0.517
CCRT/CC/RT†	27	23
Surgery + CCRT/CC/RT‡	12	13
Patient’s awareness of diagnosis
Yes	42	36	3.332	0.068
No	7	15
Activity status
No limitation	3	23	19.731	<0.001**
Considerable limitation	46	28

**Statistically significant at the 0.01 level (two tailed); †Concurrent chemotherapy + radiotherapy/chemotherapy/radiotherapy; ‡Surgery + concurrent chemotherapy + radiotherapy/chemotherapy/radiotherapy. MDE – Major depressive episode

Table 3 shows the association between clinical variables and diagnosis of MDE with melancholic features using Pearson's Chi-square test/Fisher's exact test. There was a statistically significant association between the activity status of the patient and the diagnosis of MDE with melancholic features using Fisher's exact (P = 0.019). Patients with HNC who had considerable limitation of their activity status were more likely to be diagnosed with MDE with melancholic features.

Table 3

Association between clinical variables and diagnosis of major depressive episode with melancholic features

Clinical data	MDE with melancholic features		χ2/Fisher’s exact test	P
	Present	Absent
Cancer staging
Early	5	51	1.865	0.172†
Advanced	8	36
Treatment received
Surgery	3	22	0.848	0.689‡
CCRT/CC/RT¦	8	42
Surgery + CCRT/CC/RT¶	2	23
Patient’s awareness of diagnosis
Yes	11	67	-	0.727‡
No	2	20
Activity status
No limitation	0	26	-	0.019‡,*
Considerable limitation	13	61

†Chi-square test, ‡Fisher’s exact test, *Statistically significant at the 0.05 level (two tailed). §Concurrent chemotherapy + radiotherapy/chemotherapy/radiotherapy; ¶Surgery + concurrent chemotherapy + radiotherapy/chemotherapy/radiotherapy. MDE – Major depressive episode

Table 4 shows the association between clinical variables and diagnosis of dysthymia using Fisher's exact test. There was a statistically significant association between the treatment received by the patient and the diagnosis of dysthymia using Fisher's exact test (P = 0.002). HNC patients treated with only chemotherapy or radiotherapy were less likely to be given a diagnosis of dysthymia, and patients treated with surgery in addition to chemotherapy or radiotherapy were more likely to present with dysthymia.

Table 4

Association between clinical variables and diagnosis of dysthymia

Clinical data	Dysthymia		Fisher’s exact test	P
	Present	Absent
Cancer staging
Early	5	51	-	0.745
Advanced	5	39
Treatment received
Surgery	2	23	10.963	0.002**
CCRT/CC/RT†	1	49
Surgery + CCRT/CC/RT‡	7	18
Patient’s awareness of diagnosis
Yes	7	71	-	0.687
No	3	19
Activity status
No limitation	1	25	-	0.447
Considerable limitation	9	65

**Statistically significant at the 0.01 level (two tailed); †Concurrent chemotherapy + radiotherapy/chemotherapy/radiotherapy; ‡Surgery + concurrent chemotherapy + radiotherapy/chemotherapy/radiotherapy

DISCUSSION

Individuals diagnosed with HNC are faced with many psychosocial challenges. This is one of the few studies from South India which has looked at the prevalence of depressive disorders in HNC using standardized diagnostic tool.

Studies have found that patients with HNC have some of the highest documented rates of depressive disorders.[4] In our study, we found that 49% of the patients had MDD and 13% had MDD with melancholic features. These rates are higher compared to that of previous studies looking at the prevalence of depression in cancer in general. A meta-analytic review of 66 studies done in oncological and hematological settings by Mitchell et al. showed that the prevalence of depression varied significantly.[5] They reported a meta-analytical pooled prevalence of syndromal depression of 16.3% and meta-analytical pooled prevalence of depression/adjustment disorder of 31.6%.[5] In India, the prevalence of MDD among cancer patients in different studies conducted by Alexander et al.[6] and Mendonsa and Appaya[7] was 32% and 25.7%, respectively.

However, when we look at HNC in particular, previous literature shows higher prevalence of depression, in around 15%–50% of patients across the disease trajectory, which is consistent with the findings of our study.[89] The possible factors leading to higher rates of depression as per existing research are body disfigurement, difficulty in eating and communicating, changes in sexuality, presence of comorbidities, history of substance abuse, impaired ability to work, dissatisfaction with care, and poor delivery of information by the medical team.[1011]

Our study also revealed that patients who had considerable functional limitation were more likely to have a diagnosis of MDD. This is consistent with a previous research which has shown that physical dysfunction and functional impairment are major predictors of depressive symptoms.[12]

Our study also revealed that 10% of the patient sample received a diagnosis of dysthymia. The pooled prevalence of dysthymia in a meta-analysis by Mitchell et al.[5] was 2.7%, and 12-month prevalence in a large epidemiological study by Kuhnt et al.[13] was 4.4%. However, both these studies looked at varied tumor entities and not just HNC.

Our study also showed that patients who were treated with surgery plus chemotherapy/radiotherapy were more likely to be diagnosed with dysthymia. This is understandable as surgery leads to significant changes in physical appearance; facial disfigurement; and functional impairment such as difficulty in speaking, swallowing, and breathing.[4]

Our study had few limitations such as small sample size and cross-sectional design. Furthermore, one-time assessment means that our study cannot capture the full spectrum of psychiatric manifestations seen throughout the trajectory of the disease. We did not explore how the individuals were affected in their functioning due to the cancer or its treatment. Individual factors such as personality, coping strategies, social support, and quality of life were not assessed, which could have influenced the psychiatric outcome. The study cannot be generalized to all the patients with HNC as participants were patients admitted for treatment and were likely having severe illness.

CONCLUSION

Depressive disorders are highly prevalent in HNC patients, and clinicians need to be sensitive to this. Psycho-oncological and psychotherapeutic interventions are the need of the hour and would significantly influence patient's quality of life, treatment compliance, and prognosis.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

Appendix Table 3

Association between socio-demographic variables and diagnosis of major depressive episode with melancholic features

Sociodemographic variable	MDE with melancholic features		Fisher’s exact test	P
	Present	Absent
Sex
Male	9	55	-	0.765
Female	4	32
Age
≤40	3	18	2.424	0.317
41-59	9	45
≥60	1	24
Education
Less than SSC	9	70	-	0.464
More than or equal to SSC	4	17
Marital
Married	12	78	-	1.000
Single/widow/widower	1	9
Occupation
Unemployed	4	25	-	1.000
Employed	9	62
Work
Yes	5	32	-	1.000
No	8	55
Religion
Hindu	11	60	1.682	0.426
Muslim	2	26
Christian	0	1
Socioeconomic status
Middle class	5	31	1.276	0.573
Lower middle class	6	30
Lower class	2	26

MDE – Major depressive episode