Relapse of Tobacco Consumption in Patients Treated for Head and Neck Cancer: a Cross Sectional Study

Abstract

Head and neck cancers are a major healthcare burden in the Indian Subcontinent mainly due to tobacco consumption. While advances in treatment have resulted in better control rates and survival, relapse of tobacco consumption contributes to recurrences. The authors embarked upon a cross-sectional questionnaire-based study to assess the burden of tobacco relapse along with factors resulting into relapse. Patients on follow-up after treatment of tobacco-related head and neck cancers in a dedicated cancer institute were included in the study. Demographic data was noted, and responses were obtained on a five-point Likert’s scale for reasons of relapse which were analyzed with a median and interquartile range. Predictors of relapse were assessed using multivariate logistic regression analysis. In our sample size of 500, relapse rate was 15%, majority of which occurred in the first year of follow-up when disease was loco-regionally controlled. Most common causative factor for relapse was lower motivation to quit. Care givers must incorporate tobacco cessation counseling as an integral part of head and neck cancer treatment.

Introduction

Indian subcontinent has a large burden of head and neck cancer due to rampant tobacco consumption. Since tobacco consumption incidence and pattern have been established, measures to curb it have borne fruit.[1]

Continuation of tobacco habits after diagnosis of cancer is associated with treatment-related side effect, disease recurrence, second primary cancers, and mortality.[2–4] Therefore, tertiary prevention of tobacco is vital and information about patient’s tobacco habit and factors associated with continued consumption would be important considerations.

Literature review suggests prevalence of continued tobacco use among patients with head and neck cancer to range from 20 to 50%.[5, 6] However, there is a paucity of such data from India, which is the hotbed for tobacco-related head and neck cancers particularly on factors that compel patients to continue consuming tobacco after diagnosis despite treatment-related morbidity.[7] This information is critical because it would enable focused tobacco cessation interventions.

To meet this lacuna, the authors embarked upon this observational study.

Aims and Objectives

1. To quantify relapse of tobacco habit among head neck cancer patients on follow up.

2. Objective evaluation of patient self assessment of causative factors for relapse.

Materials and Methods

This prospective observational study was conducted at a tertiary level cancer centre.

Methodology

Patients with squamous cell carcinoma of head and neck were included if they had undergone oncological treatment with curative intent. Patients who reported for follow up were served a questionnaire if they consented to go ahead with the study. Authors helped patients who were illiterate by reading out the questionnaire and obtaining verbal responses.

The patients were interviewed for maintenance of tobacco habits after treatment. Patients were asked about his/her status regarding tobacco and alcohol habits and, in case the status was abstinence, for how long.

Habit maintenance was defined as a setting in which the patient continued consumption of tobacco and/or alcohol or if he/she resumed these habits after an abstinence period. Habit cessation was considered when the patient had not consumed tobacco or alcohol for at least 30 days from the interview date.[8]

The data regarding treatment history and disease staging were collected from the patients’ medical records. The oncological status was obtained from clinical assessment carried out during the visit preceding the interview. The following data were then registered in a standard card (Fig. 1)

1. Patient’s age (at diagnosis) and gender;

2. Primary tumor site and stage (clinical stages from I to IV);

3. Type of treatment (surgical or non-surgical);

4. Tobacco and alcohol use status before the treatment (YES or NO);

5. Tobacco and alcohol use after treatment (YES or NO);

6. Oncological status: On treatment/locoregionally controlled/ Recurrence

7. Time of tobacco relapse: On treatment, < 6 months; 6 to 12 months; 12 to 24 months; > 24 months)

8. Type of tobacco relapse and reason for relapse

Fig. 1.

Proforma for data collection and questionnaire

Inclusion Criteria

Patients on follow up of tobacco-related head neck cancers (oral and laryngopharyngeal squamous cell carcinoma).

Exclusion Criteria

1. Never consumers

2. Non squamous histology.

Statistical Analysis

Sample Size

For sample size calculation, following formula was used:

N = Z²x P x (1-P)/D².

Z_{1 − α/2} = Is standard normal variate (at 5% type 1 error (P < 0.05) it is 1.96).

P = Expected proportion of tobacco habit relapse among the head and neck cancer patients was taken as 35% (Pinto FR et al.[9]).

Q = 100-P therefore Q = 100–35 = 65.

D = Absolute precision to be decided at 5%.

N = (1.96)² × 0.35 × (1–0.35)/(0.05)² = 349.58.

In order to achieve robust results from this observational study, the authors considered a sample size of 500.

After data collection, data entry was performed in MS Excel. Data analysis was performed with the help of SPSS Software version 23 (IBM, USA). Responses were analyzed from the 5 point Likert’s scale with a median and Interquartile range. Qualitative data of demographics age groups, gender, marital status, occupation, socio-economic status, type of addiction, and other variables related to disease were presented with the help of frequency and percentage table. Univariate analysis was used to demonstrate association among relapse, and causative variables were assessed with χ² test. Multivariable regression model was used to assess independent predictors of relapse of tobacco habits. Level of significance was considered at 95% confidence interval (p < 0.05).

Results

Patient Characteristics

1. Majority of the patients were in the age range of 21–70 years, predominantly males.

2. Most of the patients belonged to low socio-economic class (60.4%) and were unskilled workers.

3. Most patients had advanced staged disease (stages III and IV), with gingivobuccal complex being the most frequent site (52%).

4. Khaini and pan masala were the most common forms of tobacco consumption (68%) and 42.6% of the patients consumed both tobacco and alcohol.

Tables 1, 2, and 3 summarize the demographic data, socio-demographic data, disease status and treatment related data .

Table 1.

Demographic data (n = 500)

Variable	Frequency	Percentage value
Age
0–20 years	0	0
21–50 years	268	53.6
51–70 years	217	43.4
> 70 years	15	3
Gender
Male	457	91.4
Female	43	8.6
Marital status
Married	498	99.6
Single	2	0.4
Widow	0	0
Education
Illiterate	244	48.8
Primary	58	11.6
Upper primary	26	5.2
Secondary	59	11.8
Senior secondary	40	8
Under graduation	73	14.6
Post graduation	0	0

Table 2.

Socio-demographic data (n = 500)

Variable	Frequency	Percentage value
Occupation
Professionals	4	0.8
Semi-professionals	46	9.2
Clerical, shop owner, farmer	180	36
Skilled worker	55	11
Unskilled workers	158	31.6
Unemployed (Housewife, retired)	57	11.4
Socio-economic status
BPL	50	10
Lower class	302	60.4
Middle class	149	29.8
Upper class	0	0
Type of addiction
Khaini	237	47.4
Pan masala	103	20.6
Pan	14	2.8
Supari	15	3
Mava	1	0.2
Bidi, Cigarette	4	0.8
Others and combination of above	126	25.2
Alcohol + tobacco
Yes	213	42.6
No	287	57.4

Table 3.

Data related to disease status and treatment (n = 500)

Variable	Frequency	Percentage Value
Site of disease
Gingivo-buccal complex	391	78.2
Tongue, floor of mouth, oro-pharynx	89	17.8
Hypo-pharynx, and larynx	20	4
Stage of disease
Stage I	47	9.4
Stage II	98	19.6
Stage III	70	14
Stage IV	285	57
Treatment status
Disease controlled	300	60
On treatment	134	26.8
Residual disease/recurrence	66	13.2

Relapse of Habit

1. Seventy-five patients reported that they resumed tobacco consumption after treatment for head neck cancer. This figure represents 15% of the total study sample (n = 500).

2. Among the 75 patients who reported relapse of tobacco habit, majority occurred in less than 6 months of treatment completion (52% n = 39) (Table 4).

Table 4.

Timing of tobacco relapse

Time of relapse	Frequency	Percentage
On treatment	4	5.33
< 6 months post T/t	39	52
6 months–1 year	28	37.33
1 year–2 years	3	4
> 2 years	1	1.33
	75	100

Causative Factors

• Most common causative factor for relapse was lower quitting motivation as per responses on Likert’s scale (Table 5)

Table 5.

Likert’s scale analysis of causative factors for relapse.

	Median	Interquartile range
Lack of awareness	2	2–3
Could not quit	2	2–2
Peer pressure	2	2–3
Worthlessness	3	3–4

5-point Likert’s scale: 1 most likely; 5 least likely

In the univariate nominal logistic regression, disease status (P < 0.0001), socio-economic status (P = 0.032), and gender (P = 0.012) were predictors for relapse of tobacco habits, which were also found to be independent predictors on multivariable regression model (Table 6).

Table 6.

Summary of multivariable logistic regression analyses

Variables in the equation	B	S.E.	df	Sig.	Exp (B)
Age group	− 0.100	0.236	1	0.67161	0.905
Gender	2.095	1.032	1	0.04232	8.129
Socio-economic status (SES)	0.619	0.222	1	0.00537	1.858
Disease status	− 0.697	0.164	1	0.00002	0.498
Alcohol	0.309	0.265	1	0.24349	1.363
Constant	− 0.841	1.387	1	0.54417	0.431
a. Variable(s) entered on step 1: age group, sex, SES, current status, alcohol.

Discussion

Majority of patients in our study had advanced stage disease (71%) at the time of diagnosis. Thus, majority were subjected to multimodality treatment and suffered from incumbent morbidity of intensive treatment. Despite that our data suggests that most relapses occured within first 6 months of treatment completion implying that the urge of consumption was stronger than the will to quit. This again is borne out in the most common response by our cohort, i.e., inability to quit. Besides, a diagnosis of cancer (tobacco related) by itself is not a deterrent to tobacco consumption. A large population-based analysis from the USA demonstrates that approximately 1/3rd of patients diagnosed with cancer continue to smoke despite the knowledge that smoking is deleterious to health.[10]

Enough literature exists on the methodology to quit tobacco habit, i.e., awareness, behavioral changes, and pharmacotherapy[11]; however, intervention for tertiary prevention in these vulnerable patients would be required within the hospital while undergoing intensive cancer treatment to ensure abstinence in the post treatment period.

The authors suggest the following interventions to potentially reduce the relapse of tobacco habits:

1. Incorporation of preventive oncology specialist in the tumor board decisions. Additionally, emphasis should be made on sending patients for tobacco-related counseling during each follow up visit for at least 1 year following treatment conclusion since majority of relapses occur soon after treatment completion as has been demonstrated in our study as well as emphasized in a review by Chang et al.[11]

2. Counselors in the form of cancer survivors must be incorporated during active cancer treatment.[12]

This study is not without inherent drawbacks of a cross-sectional study. A relapse rate of 15% could be an underestimation considering it was a recorded response by the patient who might have resorted to denial of the habit fearing ire of the relatives and care givers. This could be overcome by acquiring the responses in confidentiality through patient self assessment. However, the patient profile (majority were illiterate) led the authors to obtain verbal responses after reading out the questionnaire to them.

There is scant data from India on estimates of continued tobacco consumption during cancer treatment.[7] This study is the largest, which also analyzes reasons behind relapse of the habit in a focused subsite, i.e., the head and neck known to have significant treatment-related morbidity potentially resulting into swallowing difficulty.

Conclusion

This is the largest study assessing relapse of tobacco consumption in treated head and neck cancer patients as well as reasons behind tobacco relapse. Intensive treatment on the head and neck is not a deterrent to resuming the habit with a relapse of 15%. Since majority of relapses occur soon after treatment conclusion, it is crucial that head and neck oncologists are sensitized to the complex psychic and behavioral mechanisms of tobacco habit relapse. Additionally, incorporation of preventive oncology specialists as well as cancer survivors in the form of counselors during active cancer treatment cannot be underestimated.

Compliance with Ethical Standards

Conflicting Interest

The authors declare that they have no conflict of interest.